TW201212159A - Electrostatic chuck and production method thereof - Google Patents

Electrostatic chuck and production method thereof Download PDF

Info

Publication number
TW201212159A
TW201212159A TW100128677A TW100128677A TW201212159A TW 201212159 A TW201212159 A TW 201212159A TW 100128677 A TW100128677 A TW 100128677A TW 100128677 A TW100128677 A TW 100128677A TW 201212159 A TW201212159 A TW 201212159A
Authority
TW
Taiwan
Prior art keywords
electrostatic chuck
dielectric substrate
sintered body
particle diameter
less
Prior art date
Application number
TW100128677A
Other languages
Chinese (zh)
Inventor
Kaduko Ishikawa
Junji Yonezawa
Toshihiro Aoshima
Original Assignee
Toto Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Toto Ltd filed Critical Toto Ltd
Publication of TW201212159A publication Critical patent/TW201212159A/en

Links

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
    • H01L21/683—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere for supporting or gripping
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02N—ELECTRIC MACHINES NOT OTHERWISE PROVIDED FOR
    • H02N13/00—Clutches or holding devices using electrostatic attraction, e.g. using Johnson-Rahbek effect
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q3/00—Devices holding, supporting, or positioning work or tools, of a kind normally removable from the machine
    • B23Q3/15—Devices for holding work using magnetic or electric force acting directly on the work
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q3/00—Devices holding, supporting, or positioning work or tools, of a kind normally removable from the machine
    • B23Q3/15—Devices for holding work using magnetic or electric force acting directly on the work
    • B23Q3/152—Rotary devices
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
    • H01L21/683—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere for supporting or gripping
    • H01L21/6831—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere for supporting or gripping using electrostatic chucks
    • H01L21/6833—Details of electrostatic chucks

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • General Physics & Mathematics (AREA)
  • Manufacturing & Machinery (AREA)
  • Computer Hardware Design (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Power Engineering (AREA)
  • Container, Conveyance, Adherence, Positioning, Of Wafer (AREA)
  • Jigs For Machine Tools (AREA)

Abstract

The electrostatic chuck (1) of the present invention is provided with a dielectric substrate (3) having protrusions (3a) formed on the main surface of the side for mounting an object to be attached, and planar portions (3b) formed on the periphery of the protrusions (3a). The electrostatic chuck (1) is characterised by: the dielectric substrate (3) being formed from a sintered polycrystalline ceramic body; and the interference fringe occupancy ratio, obtained using a laser microscope, being less than 1% on the main surface. Provided are an electrostatic chuck that can curb particle generation, and an electrostatic chuck production method.

Description

201212159 六、發明說明： 【發明所屬之技術領域】 本發明的形態是有關一般靜電吸盤及靜電吸盤的製造 方法。 【先前技術】 倉虫刻，在進行 CVD ( Chemical Vapor Deposition)、 濺射、離子注入、灰化、曝光、檢查等的基板處理裝置中 ，使用靜電吸盤作爲吸附保持被吸附物（半導體晶圓或玻 璃基板等）的手段。 · 在此，一旦靜電吸盤的載置面與被吸附物互相摩擦， 則恐有發生微粒之虞。並且，一旦靜電吸盤的載置面與被 吸附物的接觸面積變大，則被吸附物的吸附脫離應答性恐 有變差之虞。 因此，有在靜電吸盤的載置面側設置突起部，縮小接 觸面積，謀求微粒污染的抑制及被吸附物的吸附脫離應答 性的提升之技術爲人所知。 而且，有在靜電吸盤的載置面側設置突起部，且將突 起部的頂面拋光硏磨，而於頂面形成表面粗度Ra爲0.25 S 以下的平坦面之技術被提案（參照專利文獻1 )。 在此專利文獻1所開示的技術是在鏡面硏磨突起部的 頂面或側面及突起部的周邊的平面部（凹部的底面）之下 ，即使被吸附物的背面接觸於該等的部分’還是可抑制微 粒的發生（參照專利文獻1的[0008]、[0029]、[003 5]等） 201212159 然而，在專利文獻1所開示的技術中是利用噴沙法來 形成突起部。因此，有時在突起部的表面區域或平面部的 表面區域發生龜裂等的缺陷部。當如此的缺陷部在表面區 域內部存在時，恐有以缺陷部作爲基點，表面區域的一部 分脫離而產生微粒之虞。 尤其是近年來附著於被吸附物的背面等的微粒數的限 制有變嚴的傾向。因此，若無法使表面區域內部存在的缺 陷部減少，則恐有無法對應於微粒數的限制之虞。 如此在表面區域內部存在的缺陷部是無法直接從外部 辨識。亦即，以往對缺陷部的定量評價困難。 並且，如此在表面區域內部存在的缺陷部是專利文獻 1所開示的拋光硏磨法無法除去，使用砥石加工法、雷射 雕刻法、噴砂處理法（shot blasting )等，恐也會有缺陷 部更增加之虞》 而且’未考慮有關構成突起部的頂面或側面及突起部 的周邊的平面部之材料的結晶粒子徑，恐也會有微粒增加 之虞。 [先行技術文獻] [專利文獻] [專利文獻1]特開2003-86664號公報 【發明內容】201212159 VI. Description of the Invention: [Technical Field] The present invention relates to a method of manufacturing a general electrostatic chuck and an electrostatic chuck. [Prior Art] In a substrate processing apparatus that performs CVD (Chemical Vapor Deposition), sputtering, ion implantation, ashing, exposure, inspection, etc., an electrostatic chuck is used as an adsorption holding adsorbate (semiconductor wafer or Means for glass substrates, etc.). - Here, once the mounting surface of the electrostatic chuck and the object to be adsorbed rub against each other, there is a fear that particles will be generated. Further, when the contact area between the mounting surface of the electrostatic chuck and the object to be adsorbed becomes large, the adsorption-desorption responsiveness of the adsorbate may be deteriorated. Therefore, there is a technique in which a projection is provided on the mounting surface side of the electrostatic chuck, and the contact area is reduced, and the suppression of particulate contamination and the improvement of the adsorption/desorption responsiveness of the adsorbate are known. In addition, a technique in which a projection is provided on the mounting surface side of the electrostatic chuck, and the top surface of the projection is polished and honed, and a flat surface having a surface roughness Ra of 0.25 S or less is formed on the top surface (refer to the patent document). 1 ). The technique disclosed in Patent Document 1 is that the top surface or the side surface of the mirror honing protrusion and the flat portion (the bottom surface of the concave portion) around the protrusion portion, even if the back surface of the object to be adsorbed contacts the portions' The occurrence of fine particles can be suppressed (refer to [0008], [0029], [003 5], etc. of Patent Document 1). 201212159 However, in the technique disclosed in Patent Document 1, the projections are formed by the sandblasting method. Therefore, a defective portion such as a crack may occur in the surface region of the protruding portion or the surface region of the flat portion. When such a defect portion exists inside the surface region, it is feared that the defect portion is used as a base point, and a part of the surface region is separated to generate a flaw of the particles. In particular, in recent years, restrictions on the number of particles adhering to the back surface of the object to be adsorbed and the like tend to be severe. Therefore, if the number of defects existing in the surface region cannot be reduced, there is a fear that the number of particles cannot be limited. Thus, the defective portion existing inside the surface region cannot be directly recognized from the outside. That is, it has been difficult to quantitatively evaluate the defective portion in the past. Further, the defect portion existing in the surface region as described above cannot be removed by the polishing honing method disclosed in Patent Document 1, and the defect portion may be used by using a vermiculite processing method, a laser engraving method, a shot blasting method, or the like. Further, "there is no consideration of the crystal particle diameter of the material of the flat portion constituting the top surface or the side surface of the protrusion portion and the periphery of the protrusion portion, and there is a fear that the number of particles increases. [Prior Art Document] [Patent Document] [Patent Document 1] JP-A-2003-86664

-6 - 201212159 (發明所欲解決的課題） 本發明的形態是根據該課題的認識而硏發者，提供一 種可抑制微粒的發生之靜電吸盤及靜電吸盤的製造方法。 (用以解決課題的手段） 第1發明的靜電吸盤，係具備介電質基板，該介電質 基板係具有：形成於載置被吸附物側的主面之突起部、及 形成於上述突起部的周邊之平面部，其特徵爲： 上述介電質基板係由多結晶陶瓷燒結體所形成， 利用雷射顯微鏡所求取之上述主面的干涉條紋佔有面 積率爲未滿1 %。 若根據此靜電吸盤，則由於干涉條紋佔有面積率未滿 1%，因此可使表面區域的一部分脫離下產生的微粒數大幅 度地降低。 又，第2發明係於第1發明中，上述多結晶陶瓷燒結體 的結晶粒的平均粒子徑係比上述突起部的高度尺寸更小。 藉由此靜電吸盤，可抑制結晶粒從介電質基板脫粒。 並且，即使假設結晶粒脫粒，還是可抑制突起部的形狀變 化。 又，第3發明係於第2發明中’上述平均粒子徑爲1·5μιη 以下。 若根據此靜電吸盤，則可更確實地抑制結晶粒從介電 質基板脫粒。並且，即使假設結晶粒脫粒’還是可抑制突 起部的形狀變化。 201212159 又，第4發明係於第2發明中，上述結晶粒的粒子徑分 布的標準偏差爲Ιμιη以下。 若根據此靜電吸盤，則可更確實地抑制結晶粒從介電 質基板脫粒。並且，即使假設結晶粒脫粒，還是可抑制突 起部的形狀變化。 又，·第5發明係於第1發明中，上述介電質基板係由多 結晶氧化鋁燒結體所形成，體積密度爲3.9 6以上。 若根據此靜電吸盤，則可使成爲底層的多結晶氧化鋁 燒結體形成緻密的組織，因此可更確實地抑制結晶粒從介 電質基板脫粒。 又，第6發明係於第1發明中，上述介電質基板係由多 結晶氧化錦燒結體所形成，氧化銘含有率爲99.9 wt%以上 〇 若根據此靜電吸盤，則可使成爲底層的多結晶氧化鋁 燒結體形成緻密的組織，因此可更確實地抑制結晶粒從介 電質基板脫粒。 又，第7發明係於第1發明中，上述介電質基板係體積 電阻率在靜電吸盤的使用溫度區域中爲l〇8ncm以上' 1 0 13 Ω c m 以下。 如此的靜電吸盤是利用強生•羅貝克力（Johnsen-Rahbek力）來吸附被吸附物者。若利用強生•羅貝克力， 則吸附力會比利用庫倫力時更強’即便是如此的靜電吸盤 ，照樣可使微粒的發生大幅度減少。 又，第8發明係於第7發明中，上述介電質基板係由多 -8- 201212159 結晶氧化鋁燒結體所形成，氧化鋁含有率爲9 9 · 4 w t %以上 〇 若爲由如此高純度的氧化鋁所形成，則可抑制氧化鋁 以外的物質所造成的污染。 又，第9發明係靜電吸盤的製造方法，該靜電吸盤係 具備介電質基板，該介電質基板係具有：形成於載置被吸 附物側的主面之突起部'及形成於上述突起部的周邊之平 面部，其特徵爲： 上述介電質基板係由多結晶陶瓷燒結體所形成， 上述主面的加工係繼續至利用雷射顯微鏡所求取之上 述主面的干涉條紋佔有面積率爲未滿1 %。 若根據此靜電吸盤的製造方法，則可使干涉條紋佔有 面積率成爲未滿1%，因此可使表面區域的一部分脫離下產 生的微粒數大幅度地降低。 [發明的效果] 若根據本發明的形態，則可提供一種可抑制微粒的發 生之靜電吸盤及靜電吸盤的製造方法。 【實施方式】 以下，一面參照圖面，一面說明有關本發明的實施形 態。另外，在各圖面中，對同樣的構成要素附上同一符號 ，詳細的說明則適當省略。 圖1是用以舉例說明本實施形態的靜電吸盤的模式剖 -9 - 201212159 面圖。另外，圖1 ( a )是用以舉例說明靜電吸盤的模式剖 面圖，圖1 (b)是圖1 (a)的A部的模式擴大圖。 如圖1(a) 、 （b)所示，在靜電吸盤1設有基台2、 介電質（dielectric)基板3、電極4。 在基台2的一方的主面（電極4側的表面）形成有由無 機材料所構成的絕緣體層5。並且，介電質基板3具有：形 成於載置被吸附物側的主面（載置面側）之突起部3a、及 形成於突起部3a的周邊之平面部3b。此突起部3a的頂面是 在載置半導體晶圓等的被吸附物時成爲載置面。另外，有 關突起部3a或平面部3b的表面性狀、剖面形狀等的詳細會 在往後敘述。 並且，設有電>1 4的介電質基板3的主面與設有絕緣體 層5的基台2的主面是以絕緣性黏著劑所黏著。此絕緣性黏 著劑硬化者會成爲接合層6。 電極4與電源10a、電源10b是以電線9所連接。另外， 電線9是設成貫通基台2，但電線9與基台2是被絕緣。在圖 1所示的例子是欲使正極、負極的電極彼此鄰接來形成於 介電質基板3之所謂的雙極型靜電吸盤。但，並非限於此 ，亦可將1個的電極形成於介電質基板3之所謂的單極型靜 電吸盤，或三極型、或其他多極型。並且，電極的數量、 形狀、配置亦可適當變更。 又，以能夠貫通靜電吸盤1的方式設有貫通孔11。貫 通孔11的一端是開口於平面部3b，另一端是經由未圖示的 壓力控制手段或流量控制手段來與未圖示的氣體供給手段 -10- 201212159 連接。未圖示的氣體供給手段是供給氦氣體或氬氣體等。 而且，在形成平面部3b下所設的空間3c會成爲所被供給的 氣體的通路》空間3c彼此間是分別連通，所被供給的氣體 可全體流傳。 並且，在載置半導體晶圓等的被吸附物時支撐被吸附 物的外周部的位置配設未圖示的環狀的突起部，亦可使前 述的氣體不會漏出。而且，在設有前述氣體供給用的貫通 孔11以外的貫通孔時，是在該貫通孔的周圍配設未圖示的 環狀的突起部，亦可使前述氣體不會漏出。 如此未圖示的環狀突起部的表面性狀、剖面形狀等， 亦可與突起部3a同樣。 更可在平面部3b設置與貫通孔1 1連通之放射狀或同心 圓狀的未圖示的氣體分配溝（凹狀的溝）。若設置如此的 氣體分配溝，則可加快氣體分配速度。 基台2是例如可使用像鋁合金或銅等那樣熱傳導率高 的金屬所形成。而且，可在其內部設置流動冷卻液或加熱 液的流路8。·另外，流路8並非一定需要，但由被吸附物的 溫度控制的觀點來看，較理想是設置。 並且，設於基台2的一方的主面的絕緣體層5，例如可 由氧化鋁（Al2〇3 )或氧化釔（γ2〇3 )等的多結晶體所形 成。而且’絕緣體層5的熱傳導率是形成比接合層6的熱傳 導率更大爲理想。此情況’將絕緣體層5的熱傳導率形成 2 W/mK以上更理想。如此一來，熱傳達性要比接合層單獨 時更良好，可使被吸附物的溫度控制性與面內溫度的均一 -11 - 201212159 性更提升。 在接合層6中是提高其熱傳導率爲理想。例如，較理 想是將熱傳導率設爲lW/mK以上，更理想是設爲1.6W/mK 以上。如此的熱傳導率是例如可在矽樹脂等中添加氧化鋁 或氮化鋁作爲塡充物來取得。又，亦可以添加的比例來調 整熱傳導率。 接合層6的厚度，若考慮熱傳達性，則是儘可能薄爲 理想。另一方面，若考慮因爲基台2的熱膨脹率與介電質 基板3的熱膨脹率的差所引起的熱剪應力，造成接合層6剝 離’則接合層6的厚度是儘可能較厚爲理想。因此，接合 層6的厚度是考慮該等而設爲〇.1 mm以上、〇.3mm以下爲理 想。 介電質基板3可依靜電吸盤的各種要求來使用各種的 材料。此情況，若考慮熱傳導率、電氣絕緣的可靠度，則 以多結晶陶瓷燒結體爲理想。多結晶陶瓷燒結體是例如可 由氧化鋁、氧化釔、氮化鋁、碳化矽等所構成的多結晶陶 瓷燒結體。 介電質基板3的材料的體積電阻率在靜電吸盤的使用 溫度區域可爲l〇8Dcm以上。 另外，本說明書的體積電阻率是利用ns規格（JIS C 2141: 1 992電氣絕緣用陶瓷材料試驗方法）所示的方法來 測定的値。此時測定是可在靜電吸盤的使用溫度區域（例 如室溫（25°C程度））進行。 又，介電質基板3是由結晶粒的平均粒子徑爲0.8 μιη以 -12- 201212159 上、1.5μηι以下的多結晶陶瓷燒結體所構成者爲理想。又 ，介電質基板3是由結晶粒的平均粒子徑爲1 μιη以上、 1 ·5μηι以下的多結晶陶瓷燒結體所構成者更理想。若由結 晶粒的平均粒子徑爲0.8μιη以上、1.5μιη以下的多結晶陶瓷 燒結體所構成，則可更確實地抑制結晶粒從介電質基板3 脫粒。並且，即使假設結晶粒脫粒，還是可抑制突起部3 a 的形狀變化。另外，有關構成介電質基板3的多結晶陶瓷 燒結體的結晶粒的平均粒子徑的詳細會往後敘述。 電極4的材料可舉氧化鈦、鈦的單體或鈦與氧化鈦的 混合體、氮化鈦、碳化鈦、鎢、金、銀、銅 '鋁、鉻、鎳 、金-白金合金等。 其次，更舉例說明有關突起部3a或平面部3b的表面性 狀 '剖面形狀等。突起部3a的頂面是在載置被吸附物時成 爲載置面。因此，爲了使微粒的發生降低，以往是將突起 部的頂面設爲平坦面，進行拋光硏磨或鏡面硏磨，藉此使 不會在頂面形成微細的凹凸（例如參照專利文獻1、專利 文獻2 )。 然而’本發明者們的檢討結果明確，若將突起部的頂 面設爲平坦面，使不會在頂面形成微細的凹部，則反而微 粒數會增加。因此，在本實施形態中，將突起部3 a的頂面 3al設爲曲面的同時，在頂面3al形成微細的凹部13a (第1 凹部）（參照圖2、圖3、圖4)。 微細的凹部13&的深度是形成根據結晶粒子徑的尺寸 。此情況，微細的凹部13a的深度尺寸是3〇nm以上、 -13- 201212159 150nm以下爲理想（參照圖26)。 在此，利用噴沙法等來切削突起部3 a的部分的周邊， 藉此形成突起部3a與平面部3b的大部分的形狀。因此’在 平面部3b形成有開口於平面部3b的複數個孔3bl。並且’ 如後述般，在開口於平面部3b的孔3bl的開口周邊形成有 平坦部3 b 2 » 而且，本實施形態中，在平坦部3b2也形成微細的凹 部1 3 b (第2凹部）。 微細的凹部13b的深度尺寸是30nm以下，較理想是 20nm以下，更理想是5nm以上、20nm以下。 圖2是用以舉例說明突起部及平面部的表面性狀、剖 面形狀等的圖表。另外，圖2是利用接觸式粗度計來測定 突起部及平面部的表面者。 如圖2所示，突起部3 a的頂面3 a 1是具有朝外側突出的 曲面。而且，在突起部3 a的頂面3al形成有微細的凹部13a 〇 並且，在平面部3b設有：開口於平面部3b的複數個孔 3bl、及形成於孔3bl的開口周邊的平坦部3b2。而且，在 平坦部3b2形成有微細的凹部13b。 在此，說明有關本說明書的「頂面」。 如圖2所示，本說明書所謂的「頂面」是意指從突起 部3a的中心軸分開在L2的長度的範圍內所具有的部分。在 此，L2是突起部3a的底部的長度L1的80%的長度。 另外，只要突起部3a的頂面3al具有曲面即可，頂面-6 - 201212159 (Problems to be Solved by the Invention) An aspect of the present invention is to provide an electrostatic chuck and a method of manufacturing an electrostatic chuck capable of suppressing the occurrence of fine particles. (Means for Solving the Problem) The electrostatic chuck according to the first aspect of the invention includes a dielectric substrate having a protrusion formed on a main surface on which the object to be adsorbed is placed, and a protrusion formed on the protrusion The planar portion around the portion is characterized in that the dielectric substrate is formed of a polycrystalline ceramic sintered body, and the interference fringe occupying area ratio of the main surface obtained by a laser microscope is less than 1%. According to this electrostatic chuck, since the interference fringe occupying area ratio is less than 1%, the number of particles generated by the partial removal of the surface region can be largely reduced. According to a second aspect of the invention, in the first aspect of the invention, the average particle diameter of the crystal grains of the polycrystalline ceramic sintered body is smaller than the height of the protrusion. By this electrostatic chuck, it is possible to suppress the degranulation of crystal grains from the dielectric substrate. Further, even if the crystal grains are threshed, the shape change of the protrusions can be suppressed. In the second invention, the average particle diameter is 1·5 μm or less. According to this electrostatic chuck, it is possible to more reliably suppress the degranulation of crystal grains from the dielectric substrate. Further, even if the crystal grain threshing is assumed, the shape change of the protruding portion can be suppressed. In the second invention, the standard deviation of the particle diameter distribution of the crystal grains is Ιμηη or less. According to this electrostatic chuck, it is possible to more reliably suppress the degranulation of crystal grains from the dielectric substrate. Further, even if the crystal grains are threshed, the shape change of the protruding portion can be suppressed. According to a fifth aspect of the invention, the dielectric substrate is formed of a polycrystalline alumina sintered body, and has a bulk density of 3.9 6 or more. According to this electrostatic chuck, the polycrystalline alumina sintered body which is the underlayer can form a dense structure, so that the crystal grains can be more reliably suppressed from being degranulated from the dielectric substrate. According to a sixth aspect of the invention, the dielectric substrate is formed of a polycrystalline oxidized luminescence sintered body, and the oxidized inclusion content is 99.9 wt% or more, and according to the electrostatic chuck, the underlayer can be formed. Since the polycrystalline alumina sintered body forms a dense structure, it is possible to more reliably suppress the degranulation of the crystal grains from the dielectric substrate. According to a seventh aspect of the invention, in the first aspect of the invention, the dielectric substrate has a volume resistivity of 1 13 8 ncm or more and 1 10 13 Ω c m or less in a use temperature region of the electrostatic chuck. Such an electrostatic chuck is a person who uses the Johnsonsen-Rahbek force to adsorb the adsorbate. If you use Johnson & Loebeck force, the adsorption force will be stronger than when using Coulomb force. Even with such an electrostatic chuck, the particle size can be greatly reduced. According to a seventh aspect of the invention, the dielectric substrate is formed of a poly-8-201212159 crystalline alumina sintered body, and the alumina content is 9 9 · 4 wt % or more. When the purity of alumina is formed, contamination by substances other than alumina can be suppressed. According to a ninth aspect of the invention, in the electrostatic chuck, the electrostatic chuck includes a dielectric substrate having a protrusion formed on a main surface on which the object to be adsorbed is placed and a protrusion formed on the protrusion The planar portion around the portion is characterized in that the dielectric substrate is formed of a polycrystalline ceramic sintered body, and the processing of the main surface continues until the interference fringe occupying area of the main surface obtained by a laser microscope The rate is less than 1%. According to the method for producing an electrostatic chuck, the area ratio of the interference fringe can be made less than 1%, so that the number of particles which can be removed from a part of the surface region can be greatly reduced. [Effect of the Invention] According to the aspect of the present invention, it is possible to provide a method of manufacturing an electrostatic chuck and an electrostatic chuck which can suppress the occurrence of fine particles. [Embodiment] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings, the same components are denoted by the same reference numerals, and the detailed description is omitted as appropriate. Fig. 1 is a plan sectional view taken along the line -9 - 201212159 of the electrostatic chuck of the embodiment. 1(a) is a schematic cross-sectional view for illustrating an electrostatic chuck, and FIG. 1(b) is a schematic enlarged view of a portion A of FIG. 1(a). As shown in FIGS. 1(a) and 1(b), the electrostatic chuck 1 is provided with a base 2, a dielectric substrate 3, and an electrode 4. An insulator layer 5 made of an inorganic material is formed on one main surface (surface on the electrode 4 side) of the base 2. Further, the dielectric substrate 3 has a projection portion 3a formed on the main surface (on the mounting surface side) on which the object to be adsorbed is placed, and a flat portion 3b formed on the periphery of the projection portion 3a. The top surface of the protrusion 3a serves as a mounting surface when an object to be adsorbed such as a semiconductor wafer is placed. Further, the details of the surface properties, the cross-sectional shape, and the like of the projection 3a or the flat portion 3b will be described later. Further, the main surface of the dielectric substrate 3 provided with electricity > 14 and the main surface of the base 2 provided with the insulator layer 5 are adhered by an insulating adhesive. This insulating adhesive hardener becomes the bonding layer 6. The electrode 4 is connected to the power source 10a and the power source 10b by a wire 9. Further, the electric wire 9 is provided to penetrate the base 2, but the electric wire 9 and the base 2 are insulated. The example shown in Fig. 1 is a so-called bipolar electrostatic chuck in which the electrodes of the positive electrode and the negative electrode are adjacent to each other to form a dielectric substrate 3. However, the present invention is not limited thereto, and one electrode may be formed of a so-called unipolar electrostatic chuck of the dielectric substrate 3, or a three-pole type or other multi-pole type. Further, the number, shape, and arrangement of the electrodes can be appropriately changed. Further, the through hole 11 is provided so as to be able to penetrate the electrostatic chuck 1. One end of the through hole 11 is opened to the flat portion 3b, and the other end is connected to a gas supply means -10-201212159 (not shown) via a pressure control means or a flow rate control means (not shown). The gas supply means (not shown) is supplied with helium gas, argon gas or the like. Further, the space 3c provided under the flat portion 3b is a passage for the supplied gas. The space 3c communicates with each other, and the supplied gas can be entirely distributed. In addition, when an object to be adsorbed such as a semiconductor wafer is placed, an annular projection (not shown) is disposed at a position where the outer peripheral portion of the adsorbate is supported, and the gas described above is prevented from leaking. In addition, when a through hole other than the through hole 11 for gas supply is provided, an annular projection (not shown) is disposed around the through hole, and the gas may not leak. The surface property, the cross-sectional shape, and the like of the annular projections (not shown) may be the same as those of the projections 3a. Further, a radial or concentric circular gas distribution groove (concave groove) communicating with the through hole 1 1 may be provided in the flat portion 3b. If such a gas distribution groove is provided, the gas distribution speed can be increased. The base 2 can be formed, for example, using a metal having a high thermal conductivity such as an aluminum alloy or copper. Further, a flow path 8 through which a coolant or a heating liquid flows may be provided inside. Further, the flow path 8 is not necessarily required, but it is preferably provided from the viewpoint of temperature control of the adsorbate. Further, the insulator layer 5 provided on one main surface of the base 2 can be formed, for example, of a polycrystalline body such as alumina (Al2?3) or yttrium oxide (?2?3). Further, it is preferable that the thermal conductivity of the insulator layer 5 is larger than the heat conductivity of the bonding layer 6. In this case, it is more preferable to form the thermal conductivity of the insulator layer 5 to 2 W/mK or more. In this way, the heat transferability is better than that of the joint layer alone, and the temperature controllability of the adsorbate and the uniformity of the in-plane temperature are further improved. It is desirable to increase the thermal conductivity in the bonding layer 6. For example, it is preferable to set the thermal conductivity to 1 W/mK or more, and more preferably to 1.6 W/mK or more. Such a thermal conductivity can be obtained, for example, by adding alumina or aluminum nitride as a ruthenium to a ruthenium resin or the like. Also, the thermal conductivity can be adjusted by the ratio added. The thickness of the bonding layer 6 is preferably as thin as possible in consideration of heat transfer property. On the other hand, considering the thermal shear stress caused by the difference between the thermal expansion coefficient of the base 2 and the thermal expansion coefficient of the dielectric substrate 3, the bonding layer 6 is peeled off, and the thickness of the bonding layer 6 is as thick as possible. . Therefore, the thickness of the bonding layer 6 is considered to be 〇1 mm or more and 〇.3 mm or less in consideration of these. The dielectric substrate 3 can use various materials depending on various requirements of the electrostatic chuck. In this case, it is preferable to use a polycrystalline ceramic sintered body in consideration of thermal conductivity and reliability of electrical insulation. The polycrystalline ceramic sintered body is, for example, a polycrystalline ceramic sintered body composed of alumina, yttria, aluminum nitride, tantalum carbide or the like. The volume resistivity of the material of the dielectric substrate 3 may be l〇8 Dcm or more in the use temperature region of the electrostatic chuck. In addition, the volume resistivity of this specification is a enthalpy measured by the method shown by the ns specification (JIS C 2141: 1 992 ceramic material test method for electrical insulation). The measurement at this time can be carried out in the use temperature region of the electrostatic chuck (e.g., room temperature (about 25 ° C)). Further, the dielectric substrate 3 is preferably composed of a polycrystalline ceramic sintered body having an average particle diameter of crystal grains of 0.8 μm and -12-201212159 and 1.5 μm or less. Further, the dielectric substrate 3 is preferably composed of a polycrystalline ceramic sintered body having an average particle diameter of crystal grains of 1 μm or more and 1·5 μηι or less. When the average particle diameter of the crystal grains is 0.8 μm or more and 1.5 μm or less, the crystal grains can be more reliably suppressed from being threshed from the dielectric substrate 3. Further, even if the crystal grains are threshed, the shape change of the protrusions 3 a can be suppressed. The details of the average particle diameter of the crystal grains of the polycrystalline ceramic sintered body constituting the dielectric substrate 3 will be described later. The material of the electrode 4 may be a monomer of titanium oxide or titanium, a mixture of titanium and titanium oxide, titanium nitride, titanium carbide, tungsten, gold, silver, copper 'aluminum, chromium, nickel, gold-platinum alloy or the like. Next, the surface property "cross-sectional shape" of the projection 3a or the flat portion 3b will be described as an example. The top surface of the projection 3a is a mounting surface when the object to be adsorbed is placed. Therefore, in order to reduce the occurrence of fine particles, conventionally, the top surface of the protrusion is a flat surface, and polishing or mirror honing is performed, so that fine unevenness is not formed on the top surface (for example, refer to Patent Document 1 Patent Document 2). However, as a result of the review by the inventors, it is clear that if the top surface of the protrusion is a flat surface so that a fine recess is not formed on the top surface, the number of fine particles increases. Therefore, in the present embodiment, the top surface 3a1 of the protrusion portion 3a is formed into a curved surface, and the fine concave portion 13a (first concave portion) is formed on the top surface 3a1 (see Figs. 2, 3, and 4). The depth of the fine recesses 13 & is a size which is formed according to the diameter of the crystal particles. In this case, the depth of the fine recessed portion 13a is preferably 3 〇 nm or more and -13 - 201212159 150 nm or less (see FIG. 26). Here, the periphery of the portion of the projection 3a is cut by a sandblasting method or the like, whereby the shape of most of the projection 3a and the flat portion 3b is formed. Therefore, a plurality of holes 3b1 opening to the flat portion 3b are formed in the flat portion 3b. Further, as will be described later, the flat portion 3b 2 is formed around the opening of the hole 3b1 that is opened in the flat portion 3b. Further, in the present embodiment, the fine recessed portion 13b (second recessed portion) is also formed in the flat portion 3b2. . The depth of the fine recessed portion 13b is 30 nm or less, preferably 20 nm or less, more preferably 5 nm or more and 20 nm or less. Fig. 2 is a graph for exemplifying the surface properties, the cross-sectional shape, and the like of the projections and the flat portions. Further, Fig. 2 shows the surface of the projection and the flat portion measured by a contact type roughness meter. As shown in Fig. 2, the top surface 3 a 1 of the projection 3 a has a curved surface that protrudes outward. Further, a fine recess 13a is formed in the top surface 3a1 of the projection 3a, and a plurality of holes 3b1 opening in the flat portion 3b and a flat portion 3b2 formed in the periphery of the opening of the hole 3b1 are provided in the flat portion 3b. . Further, a fine recess 13b is formed in the flat portion 3b2. Here, the "top surface" of this specification will be described. As shown in Fig. 2, the term "top surface" as used herein means a portion which is separated from the central axis of the projection portion 3a within the range of the length of L2. Here, L2 is a length of 80% of the length L1 of the bottom of the projection 3a. In addition, as long as the top surface 3a1 of the protrusion 3a has a curved surface, the top surface

S -14 - 201212159 3 a 1的外側可爲曲面，或直線狀的面。 其次，說明有關本說明書的曲面的「曲率半徑R」。 如圖2.所示，將頂面3 a 1的兩端部分的位置設爲P 1、P 3 ，將頂面3al的中心位置（頂面3al與突起部3a的中心軸的 交點位置）設爲P2。通過PI、P2、P3的圓的半徑爲本說明 書的曲面的「曲率半徑R」。 另外，通過PI、P2、P3的圓的中心位置是連結P1與P2 的線分的垂直二等分線、及連結P3與P2的線分的垂直二等 分線之交點。因此，從PI、P2、P3的位置求取通過PI、P2 、P 3的圓的中心位置，且求取從圓的中心位置到P 1、P 2、 P3的任一的距離，藉此可取得曲面的「曲率半徑R」。 若根據本發明者們所取得的見解，則頂面3 a 1的曲率 半徑R是比因吸附力而彎曲的板狀的被吸附物的彎曲曲線 的曲率半徑小爲理想。 如此一來，可使頂面3 al的形狀成爲對應於板狀的被 吸附物被靜電吸附時的彎曲形狀。因此，可使頂面3al與 被吸附物的背面的接觸部分的面壓降低，所以可抑制微粒 的發生。 此情況，若將曲率半徑R設爲20毫米以下，則可使頂 面3 a 1的曲率半徑形成比因吸附力而彎曲的板狀的被吸附 物的彎曲曲線的曲率半徑小。 其次，說明有關形成於頂面3al、平坦部3b2的微細的 凹部。 圖3是用以舉例說明形成於突起部的頂面之微細的凹 -15- 201212159 部的雷射顯微鏡照片。 圖4是用以舉例說明形成於平坦部的微細的凹部的掃 描型電子顯微鏡照片。 圖5是用以舉例說明將頂面3al設爲平坦面時的雷射顯 微鏡照片。 如圖3所示，在突起部3 a的頂面3 a 1形成有微細的凹部 1 3 a ° 又，如圖4所示，在平坦部3b2形成有微細的凹部13b 〇 又，圖5所示者的情況，是在頂面3al未形成有微細的 凹部13a。 在圖3、圖4所舉例說明者的情況，是頂面3al成爲曲 面，且形成有凹部13a。因此，可使頂面3al與被吸附物的 背面的接觸部分的面積大幅度地減少。亦可使微細的異物 捕捉於凹部13a的內部。 相對的，在圖5所舉例說明者的情況，是在頂面3 a 1未 形成有微細的凹部13a，因此頂面3al與被吸附物的背面的 接觸部分的面積會變大。而且，亦無法捕捉微細的異物。 又，因爲在平坦部3b2形成有凹部13b，所以即使假設 被吸附物彎曲而被吸附物的背面與平面部3b接觸，還是可 使接觸部分的面積大幅度地減少。亦可使微細的異物捕捉 於凹部13b的內部。 亦即，因爲可使與被吸附物的背面的接觸部分的面積 減少，所以可抑制微粒的發生。並且，在使微細的異物捕S -14 - 201212159 The outer side of 3 a 1 can be a curved surface or a linear surface. Next, the "curvature radius R" of the curved surface of this specification will be described. As shown in Fig. 2, the positions of the end portions of the top surface 3 a 1 are P 1 and P 3 , and the center position of the top surface 3 a (the intersection of the top surface 3 a1 and the central axis of the projection 3 a) is set. For P2. The radius of the circle passing through PI, P2, and P3 is the "curvature radius R" of the curved surface of the book. Further, the center position of the circle passing through PI, P2, and P3 is the intersection of the vertical bisector of the line connecting P1 and P2 and the vertical bisector of the line connecting P3 and P2. Therefore, the center positions of the circles passing through PI, P2, and P3 are obtained from the positions of PI, P2, and P3, and the distance from the center position of the circle to any of P1, P2, and P3 can be obtained. Get the "curvature radius R" of the surface. According to the findings obtained by the present inventors, the radius of curvature R of the top surface 3 a 1 is preferably smaller than the radius of curvature of the curved curve of the plate-shaped object to be adsorbed which is bent by the adsorption force. In this way, the shape of the top surface 3 a can be made to correspond to the curved shape when the plate-like adsorbate is electrostatically adsorbed. Therefore, the surface pressure of the contact portion between the top surface 3a1 and the back surface of the object to be adsorbed can be lowered, so that the occurrence of fine particles can be suppressed. In this case, when the radius of curvature R is 20 mm or less, the radius of curvature of the top surface 3 a 1 can be made smaller than the radius of curvature of the curved curve of the plate-shaped object to be adsorbed which is bent by the adsorption force. Next, a fine concave portion formed on the top surface 3a1 and the flat portion 3b2 will be described. Fig. 3 is a photograph of a laser microscope for exemplifying a fine concave -15 - 201212159 portion formed on the top surface of the projection. Fig. 4 is a scanning electron micrograph showing an example of a fine concave portion formed in a flat portion. Fig. 5 is a photograph showing a laser microscope when the top surface 3a1 is set to a flat surface. As shown in Fig. 3, a fine recessed portion 13a is formed on the top surface 3a1 of the projection 3a. Further, as shown in Fig. 4, a fine recessed portion 13b is formed in the flat portion 3b2. In the case of the present invention, the fine concave portion 13a is not formed on the top surface 3a1. In the case of the examples illustrated in Figs. 3 and 4, the top surface 3a1 is a curved surface, and the concave portion 13a is formed. Therefore, the area of the contact portion between the top surface 3a1 and the back surface of the object to be adsorbed can be greatly reduced. It is also possible to trap fine foreign matter inside the recess 13a. On the other hand, in the case of the example illustrated in Fig. 5, since the fine recessed portion 13a is not formed in the top surface 3a1, the area of the contact portion between the top surface 3al and the back surface of the object to be adsorbed becomes large. Moreover, it is impossible to capture fine foreign matter. Further, since the concave portion 13b is formed in the flat portion 3b2, even if the object to be adsorbed is bent and the back surface of the object to be adsorbed is in contact with the flat portion 3b, the area of the contact portion can be greatly reduced. It is also possible to trap fine foreign matter inside the recess 13b. That is, since the area of the contact portion with the back surface of the object to be adsorbed can be reduced, the occurrence of fine particles can be suppressed. And, in the fine foreign matter

-16- 201212159 捉於凹部13a及凹部13b的內部之下，可抑制微粒的發生。 表1、表2是用以舉例說明抑制微粒的發生之效果者。 另外，表1是在圖3、圖4所舉例說明者的情況，表2是 在圖5所舉例說明者的情況。 另外，表1、表2是被吸附物爲半導體晶圓，按微粒的 粒徑來總計附著於半導體晶圓的背面的微粒數者。 表1、表2中的微粒數是計測預定面積的微粒數，將其 値換算成直徑3 00mm的半導體晶圓的微粒數。 [表1] 0.15 〜0.2μιη 0.2 〜0.3μΓη 0.3 〜0.5μπι 0·5μηι以上 total 洗淨後 50 56 27 79 212 10次吸附後 25 18 25 47 115 100次吸附後 36 29 32 58 155 200次吸附後 18 27 20 23 88 300次吸附後 29 16 14 16 75 400次吸附後 23 16 14 16 69 500次吸附後 25 11 14 14 64 [表2] 0·15 〜0.2μιη 0·2 〜0·3μιη 0.3 〜0.5μηι 0.5μηι以上 total 洗淨後 79 56 9 266 410 1次吸附後 232 83 23 387 725 5次吸附後 140 45 11 263 459 10次吸附後 122 56 9 257 444 15次吸附後 140 59 14 189 402 -17- 201212159 由表1可知，在圖3、圖4所舉例說明那樣形成有微細 的凹部時，清掃靜電吸盤表面，然後重複半導體晶圓的吸 附，亦可抑制微粒的發生。 並且，形成於頂面3al的凹部13a的深度尺寸是比形成 於平坦部3b2的凹部13b的深度尺寸更大。 而且，凹部13a、凹部13b是面積廣深度淺，凹部13a 、凹部13b的側面是形成斜面。 因此，容易除去被捕捉於凹部13a、凹部13b的內部的 異物。亦即，即使在靜電吸盤表面附著有異物，還是可容 易使靜電吸盤表面的清淨狀態恢復。 相對的，由表2可知，在圖5所舉例說明那樣未形成有 微細的凹部時，清掃靜電吸盤表面，然後重複半導體晶圓 的吸附，微粒數會維持多數不變。 另外，有關凹部13a、凹部13b的深度尺寸或側面的形 狀等的詳細會在往後敘述。 表3、表4是用以舉例說明靜電吸盤表面的清淨狀態的 恢復。 另外，表3是在圖3、圖4所舉例說明者的情況，表4是 在圖5所舉例說明者的情況。 表3、表4是被吸附物爲半導體晶圓，按微粒的粒徑來 總計附著於半導體晶圓的背面的微粒數者。 表3、表4中的微粒數是計測預定面積的微粒數，將其 値換算成直徑300mm的半導體晶圓的微粒數。 並且，「初期狀態」是在靜電吸盤表面附著有異物的 -18- 201212159 狀態下使半導體晶圓吸附的情況。又，「No.l〜No.5」是 在清掃靜電吸盤表面之後使吸附於半導體晶圓的情況。另 外，清掃是以含有機溶劑的不織布來擦拭靜電吸盤表面。 [表3] 0.15 〜0·2μιη 0·2〜0·3μιη 0,3 〜〇.5μιη 0.5μηι以上 total 初期狀態 2455 2441 10676 15784 31356 No.l 47 27 54 63 191 No.2 56 25 36 29 146 No.3 34 29 34 32 129 No.4 25 25 27 23 100 No.5 11 14 9 11 45 [表4] 0.15 〜0.2μιη 0.2 〜0.3μπι 0·3 〜0·5μτη 0.5μηι以上 total 初期狀態 1577 2084 9295 16135 29091 No.l 146 79 54 303 582 No.2 101 70 41 299 511 No.3 124 77 47 266 514 No.4 100 69 40 200 409 No.5 90 77 66 184 417 在圖3 '圖4所舉例說明那樣形成有微細的凹部時，由 表3的No. 1可知，即使是以含有機溶劑的不織布來擦拭靜 電吸盤表面的程度的清掃，還是可使附著於半導體晶圓的 背面的微粒數大幅度地減少。這意味即是在靜電吸盤表面 有異物附著，還是可容易使靜電吸盤表面的清淨狀態恢復 -19· 201212159 並且’凹部13a的深度尺寸及凹部13b的深度尺寸是比 構成介電質基板3的多結晶陶瓷燒結體的結晶粒的平均粒 子徑更小。 藉此，可抑制微粒的發生，且靜電吸盤表面的清淨狀 態的恢復更容易。 利用後述的CMP法來形成以下說明的微細的凹部丨3 a 、1 3b ° 圖ό是用以舉例說明形成於頂面3al的凹部i3a的形狀 的圖。圖6(a)是凹部13a的3次元畫像，圖6(b) 、 （c )是用以舉例說明凹部13a的斷面的圖。 圖7是用以舉例說明形成於平坦部3b2的凹部13b的形 狀的圖。圖7(a)是凹部13b的3次元畫像，圖7(b)、（ c)是用以舉例說明凹部13b的斷面的圖。 如圖6所示’凹部1 3 a的側面是形成斜面，且凹部1 3 & 的底面與凹部13a的側面所成的角度（斜面的角度）是开^ 成鈍角。凹部13a的側面與頂面3al相交的部分、及 13a的側面與凹部lh的底面相交的部分是形成帶連續性的 圓之形狀。 如圖7所示，凹部13b的側面是形成斜面，凹部13b的 底面與凹部13b的側面所成的角度（斜面的角度、θ 又；疋形成 鈍角。凹部1 3 b的側面與平坦部3 b 2相交的部分、及凹1 13b的側面與凹部13b的底面相交的部分是形成帶連續性的 圓之形狀。 另外，在本說明書中所謂鈍角是意指比9 〇度大，& 201212159 180度小的角度。 又，所謂帶連續性的圓之形狀是意指利用後述的CMP 法時被化學性地侵蝕下成圓角，凹部13a的側面與頂面3al 相交的部分、凹部13a的側面與凹部13a的底面相交的部分 、凹部13b的側面與平坦部3b2相交的部分、凹部13b的側 面與凹部1 3b的底面相交的部分會順暢地連接的狀態。 因此，可消除在使靜電吸盤表面清淨化形成暗處的部 分，所以可更確實且容易地進行靜電吸盤表面的清淨狀態 的恢復。 亦即，深度淺的凹部13a、凹部13b的側面部分是形成 連續性順暢的形狀，所以可擴大與不織布等的清掃用具的 接觸面積。因此，即使是以含有機溶劑的不織布來擦拭的 程度的清掃，照樣可以順利地除去微小的異物。 相對的，由表4可知，在圖5所舉例說明那樣未形成有 微細的凹部時，以含有機溶劑的不織布來擦拭靜電吸盤表 面的程度的清掃是無法使微粒數大幅度地減少。 並且，在本實施形態的靜電吸盤1中，突起部3a的側 面與頂面3al柑交的部分、及突起部3a的側面與平面部3b 相交的部分是形成帶連續性的圓之形狀。亦即，突起部3 a 的側面與頂面3al爲曲面流暢地連接。並且，突起部33的 側面與平面部3 b爲曲面流暢地連接。 (CMP法的加工） 利用拋光硏磨法、砥石加工法、雷射雕刻法、噴砂處 -21 - 201212159 理法、噴沙法等的機械性加工法是無法將具有以上說明那 樣形狀的凹部13a或凹部13b形成於頂面3al或平坦部3b2。 並且，利用該等的機械性加工法是無法形成具有以上說明 那樣形狀的突起部3a。 以下，說明有關突起部3a、平面部3b、平坦部3b2、 孔3bl、凹部丨3a、凹部13 b等的形成方法。 首先，形成突起部3a、平面部3b的大部分的形狀。 例如，以成爲突起部3a的部分爲遮罩，利用噴沙法等 來切削未被遮罩的部分，而形成突起部3 a及平面部3b的大 部分的形狀。此時，在平面部3b形成有開口於平面部3b的 複數個孔3bl。若形成如此的孔3bl，則可使異物捕捉於複 數的孔3bl，因此可抑制微粒的發生。 此情況，孔3b 1的深度尺寸是比後述的多結晶陶瓷燒 結體的結晶粒的平均粒子徑（〇.8μπι以上、1·5μιη以下）小 爲理想。若成爲如此的淺孔，則被捕捉於孔3b 1的異物的 除去容易。另外，有關多結晶陶瓷燒結體的結晶粒的平均 粒子徑的詳細會在往後敘述。 圖8是用以舉例說明開口於平面部3b的孔3bl的深度尺 寸的圖表。 如圖8所示，孔3bl的深度尺寸是形成未滿1μ„ι，即使 從多結晶陶瓷燒結體脫粒的結晶粒進入孔3 b 1的內部，還 是可容易除去。另外，孔3bl的深度尺寸可藉由噴沙法等 的製程條件（例如所使用的硏磨材的大小等）來控制。 其次，除去遮罩，將突起部3 a加工成前述的形狀。並-16- 201212159 Under the inside of the recessed portion 13a and the recessed portion 13b, the occurrence of fine particles can be suppressed. Tables 1 and 2 are examples for illustrating the effect of suppressing the occurrence of fine particles. Further, Table 1 is the case illustrated in Figs. 3 and 4, and Table 2 is the case illustrated in Fig. 5. Further, in Tables 1 and 2, the object to be adsorbed is a semiconductor wafer, and the number of particles adhering to the back surface of the semiconductor wafer is totaled in accordance with the particle diameter of the particles. The number of particles in Tables 1 and 2 is the number of particles of a predetermined area, and the number of particles is converted into a number of particles of a semiconductor wafer having a diameter of 300 mm. [Table 1] 0.15 〜0.2μιη 0.2 〜0.3μΓη 0.3 〜0.5μπι 0·5μηι or more total After washing 50 56 27 79 212 After 10 adsorptions 25 18 25 47 115 After 100 adsorptions 36 29 32 58 155 200 adsorptions After 18 27 20 23 88 300 times after adsorption 29 16 14 16 75 400 times after adsorption 23 16 14 16 69 500 times after adsorption 25 11 14 14 64 [Table 2] 0·15 ～0.2μιη 0·2 ～0·3μιη 0.3~0.5μηι 0.5μηι or more total after washing 79 56 9 266 410 1 time after adsorption 232 83 23 387 725 5 times after adsorption 140 45 11 263 459 10 times after adsorption 122 56 9 257 444 15 times after adsorption 140 59 14 189 402 -17-201212159 It can be seen from Table 1 that when the fine concave portion is formed as illustrated in FIGS. 3 and 4, the surface of the electrostatic chuck is cleaned, and the adsorption of the semiconductor wafer is repeated, and the occurrence of fine particles can be suppressed. Further, the depth dimension of the concave portion 13a formed in the top surface 3a1 is larger than the depth dimension of the concave portion 13b formed in the flat portion 3b2. Further, the concave portion 13a and the concave portion 13b are shallow in area and the side surfaces of the concave portion 13a and the concave portion 13b are formed as inclined surfaces. Therefore, it is easy to remove foreign matter trapped inside the concave portion 13a and the concave portion 13b. That is, even if foreign matter adheres to the surface of the electrostatic chuck, the clean state of the surface of the electrostatic chuck can be easily recovered. On the other hand, as is clear from Table 2, when the fine concave portion is not formed as illustrated in Fig. 5, the surface of the electrostatic chuck is cleaned, and the adsorption of the semiconductor wafer is repeated, and the number of particles is maintained largely unchanged. The details of the depth dimension or the shape of the side surface of the recessed portion 13a and the recessed portion 13b will be described later. Tables 3 and 4 are diagrams for illustrating the recovery of the clean state of the surface of the electrostatic chuck. Further, Table 3 is the case illustrated in Figs. 3 and 4, and Table 4 is the case illustrated in Fig. 5. Tables 3 and 4 show that the adsorbed material is a semiconductor wafer, and the total number of particles adhering to the back surface of the semiconductor wafer is determined by the particle diameter of the fine particles. The number of particles in Tables 3 and 4 is the number of particles of a predetermined area, and the number of particles is converted into a number of particles of a semiconductor wafer having a diameter of 300 mm. In addition, the "initial state" is a case where the semiconductor wafer is adsorbed in the state of -18-201212159 in which foreign matter adheres to the surface of the electrostatic chuck. Further, "No. 1 to No. 5" is a case where the surface of the electrostatic chuck is cleaned and then adsorbed on the semiconductor wafer. In addition, the cleaning is performed by wiping the surface of the electrostatic chuck with a non-woven fabric containing an organic solvent. [Table 3] 0.15 〜0·2μιη 0·2~0·3μιη 0,3 〇.5μιη 0.5μηι or more total initial state 2455 2441 10676 15784 31356 No.l 47 27 54 63 191 No.2 56 25 36 29 146 No.3 34 29 34 32 129 No.4 25 25 27 23 100 No.5 11 14 9 11 45 [Table 4] 0.15 〜0.2μιη 0.2 〜0.3μπι 0·3 〜0·5μτη 0.5μηι or more total Initial state 1577 2084 9295 16135 29091 No.l 146 79 54 303 582 No.2 101 70 41 299 511 No.3 124 77 47 266 514 No.4 100 69 40 200 409 No.5 90 77 66 184 417 In Figure 3 'Figure 4 When the fine concave portion is formed as an example, it can be seen from No. 1 of Table 3 that even if the surface of the electrostatic chuck is wiped by a non-woven fabric containing an organic solvent, the particles adhering to the back surface of the semiconductor wafer can be attached. The number is greatly reduced. This means that foreign matter adheres to the surface of the electrostatic chuck, or the clean state of the surface of the electrostatic chuck can be easily restored -19·201212159 and the depth dimension of the recess 13a and the depth of the recess 13b are larger than those of the dielectric substrate 3. The average particle diameter of the crystal grains of the crystalline ceramic sintered body is smaller. Thereby, the occurrence of fine particles can be suppressed, and the recovery of the clean state of the surface of the electrostatic chuck can be made easier. The fine recessed portions a3 a and 1 3b° described below are formed by a CMP method to be described later. The figure is a view for exemplifying the shape of the concave portion i3a formed on the top surface 3a1. Fig. 6(a) is a three-dimensional image of the concave portion 13a, and Figs. 6(b) and (c) are views for explaining a cross section of the concave portion 13a. Fig. 7 is a view for exemplifying the shape of the concave portion 13b formed in the flat portion 3b2. Fig. 7(a) is a three-dimensional image of the concave portion 13b, and Figs. 7(b) and (c) are views for explaining a cross section of the concave portion 13b. As shown in Fig. 6, the side surface of the concave portion 13 3a is formed as a slope, and the angle formed by the bottom surface of the concave portion 1 3 & and the side surface of the concave portion 13a (the angle of the inclined surface) is an obtuse angle. The portion where the side surface of the recessed portion 13a intersects the top surface 3al and the portion where the side surface of the recess 13a intersects with the bottom surface of the recessed portion lh are in the shape of a circle having continuity. As shown in Fig. 7, the side surface of the recessed portion 13b is a sloped surface, and the angle between the bottom surface of the recessed portion 13b and the side surface of the recessed portion 13b (the angle of the sloped surface, θ; the 疋 forms an obtuse angle. The side surface of the recessed portion 13 b and the flat portion 3 b The portion where the intersection of 2 and the side of the concave portion 13b intersect with the bottom surface of the concave portion 13b is in the shape of a circle forming continuity. In addition, in the present specification, the obtuse angle means that it is larger than 9 turns, & 201212159 180 In addition, the shape of the circle having continuity is a portion which is chemically eroded and rounded when the CMP method described later is used, a portion where the side surface of the concave portion 13a intersects with the top surface 3al, and a side surface of the concave portion 13a. A portion intersecting the bottom surface of the concave portion 13a, a portion where the side surface of the concave portion 13b intersects with the flat portion 3b2, and a portion where the side surface of the concave portion 13b intersects with the bottom surface of the concave portion 13b is smoothly connected. Therefore, the surface of the electrostatic chuck can be eliminated. Since the cleaned portion forms a dark portion, the cleaned state of the surface of the electrostatic chuck can be recovered more reliably and easily. That is, the shallow portion 13a and the side portion of the recess 13b form continuity. Since it has a smooth shape, the contact area with a cleaning tool such as a non-woven fabric can be enlarged. Therefore, even if it is cleaned by a non-woven fabric containing an organic solvent, it is possible to smoothly remove minute foreign matter. In the case where the fine recessed portion is not formed as illustrated in Fig. 5, the cleaning of the surface of the electrostatic chuck by the non-woven fabric containing the organic solvent cannot significantly reduce the number of particles. Further, the electrostatic chuck of the present embodiment is also provided. In the first portion, the portion where the side surface of the projection 3a intersects the top surface 3a and the portion where the side surface of the projection 3a intersects the flat portion 3b is in the shape of a circle having continuity. That is, the side surface of the projection 3a is The top surface 3a1 is smoothly connected to the curved surface, and the side surface of the protruding portion 33 and the flat surface portion 3b are smoothly connected to the curved surface. (Processing by CMP method) Polishing honing method, vermiculite processing method, laser engraving method, sand blasting place -21 - 201212159 The mechanical processing method such as the method and the sandblasting method cannot form the concave portion 13a or the concave portion 13b having the shape described above on the top surface 3a1 or the flat portion 3b2. Further, the above-described mechanical processing method cannot form the projection 3a having the shape as described above. Hereinafter, the projection 3a, the flat portion 3b, the flat portion 3b2, the hole 3bl, the recess 丨3a, the recess 13b, and the like will be described. First, the shape of most of the projection 3a and the flat portion 3b is formed. For example, a portion that is the projection 3a is used as a mask, and a portion that is not covered is cut by a sandblasting method or the like to form a projection. The shape of most of the portion 3 a and the flat portion 3b. In this case, a plurality of holes 3b1 opening in the flat portion 3b are formed in the flat portion 3b. When such a hole 3b1 is formed, foreign matter can be trapped in the plurality of holes 3b1. Therefore, the occurrence of particles can be suppressed. In this case, the depth of the pores 3b 1 is preferably smaller than the average particle diameter (〇8 μm or more and 1·5 μmη or less) of the crystal grains of the polycrystalline ceramic sintered body to be described later. When such a shallow hole is formed, the removal of the foreign matter caught in the hole 3b 1 is easy. Further, the details of the average particle diameter of the crystal grains of the polycrystalline ceramic sintered body will be described later. Fig. 8 is a graph for illustrating the depth dimension of the hole 3b1 opened to the flat portion 3b. As shown in Fig. 8, the depth dimension of the hole 3b1 is less than 1 μm, and even if the crystal grain which is threshed from the polycrystalline ceramic sintered body enters the inside of the hole 3b1, it can be easily removed. In addition, the depth dimension of the hole 3b1 It can be controlled by process conditions such as a sandblasting method (for example, the size of the honing material to be used, etc.) Next, the mask is removed, and the protrusion 3a is processed into the aforementioned shape.

S -22- 201212159 且，此時，在開口於平面部3b的複數個孔3bl的開口 形成平坦部3b2。而且，在突起部3a的頂面3al形成前 細的凹部13a的同時，在平坦部3b2形成前述微細的 13b ° 此情況，根據本發明者們所取得的見解，若利用 (Chemical Mechanical Polishing;化學機械硏磨） 則可一次形成前述突起部3a、平坦部3b2、凹部13a、 1 3b ° CMP法是在一般進行平坦化加工時使用。因此， 像無法形成具有前述那樣的形狀的突起部3a，更不用 法形成微細的凹部13a、凹部13b。 然而，根據本發明者們所取得的見解，若利用含 磨液（泥漿（sluury ))的化學成分的作用，則可形 細的凹部13a、凹部13b。 亦即，利用多結晶陶瓷燒結體所具有對蝕刻速度 晶面方位依存性，可形成微細的凹部13a、凹部13b。 ，在多結晶陶瓷燒結體的表面區域中，成爲容易被蝕 結晶面方位的部分會先被蝕刻，因此可形成微細的 1 3a、凹部 1 3b。 並且，可藉由含於硏磨液的砥粒之機械性的硏磨 、及含於硏磨液的化學成分之化學性的硏磨效果來形 起部3a、平坦部3b2。此情況，平坦部3b2是形成於？ 的周邊。 在此，舉例說明CMP法的製程條件。 周邊 述微 凹部 CMP 法， 凹部 可想 說無 於硏 成微 的結 亦即 刻的 凹部 效果 成突 ,3b 1 -23- 201212159 硏磨布，例如可爲硬質發泡聚氨酯硏磨布等》硏磨盤 的旋轉速度可設爲60rpm，荷重負荷可設爲0.2kg/cm2等。 含於硏磨液的砥粒可由Si02 (氧化矽）、Ce02 (氧化鈽） 、Ti02 (氧化鈦）、MgO (氧化鎂）、Y2〇3 (氧化釔）、 Sn02 (氧化錫）等所構成者。並且，砥粒對硏磨液的比例 可設爲10〜20wt%程度。含於硏磨液的化學成分可爲pH調 整劑、砥粒的分散劑、界面活性劑等。此情況，若考慮前 述的結晶異方性蝕刻，則最好硏磨液爲鹼性。因此，硏磨 液的氫離子指數是pH8〜13程度。另外，硏磨液的供給量 ，例如可爲20cc /分程度。 而且，根據本發明者們所取得的見解，加工時間爲重 要的要素。 亦即，若加工時間短，則成爲平坦化加工，無法形成 具有前述那樣形狀的突起部3a，且亦無法形成微細的凹部 1 3a、凹部1 3b。例如，數分鐘程度的加工時間是平坦化加 工。 另一方面，若爲數小時程度的加工時間，則可形成具 有前述那樣形狀的突起部3 a，且亦可形成微細的凹部1 3 a 、凹部1 3 b。 又，由於突起部3a的頂面3al比平面部3b更容易加工 ，因此可構成前述凹部13a的深度尺寸與凹部13b的深度尺 寸的關係。亦即，形成於頂面3al的凹部13a的深度尺寸可 比形成於平坦部3b2的凹部13b的深度尺寸更大。 另外，前述的加工時間可按照CMP法的其他製程條件 • 24- 201212159 (例如硏磨液的氫離子指數等）來適當變更β 又’亦可考慮後述的干涉條紋佔有面積率。亦即，不 僅凹部13a或凹部13b的形成，可至後述的干涉條紋佔有面 積率形成未滿1 %爲止進行CMP法的加工。另外，有.關干涉 條紋佔有面積率等的詳細會在往後敘述。 又’突起部3 a的高度尺寸可比後述多結晶陶瓷燒結體 的結晶粒的平均粒子徑大。或’多結晶陶瓷燒結體的結晶 粒的平均粒子徑可比突起部3 a的高度尺寸更小。 若這樣形成’則可抑制結晶粒從介電質基板3脫粒。 並且’即使假設結晶粒脫粒，還是可抑制突起部3a的形狀 變化。 圖9是用以舉例說明微細的凹部的長度測定之雷射顯 微鏡照片。 圖1 〇是用以舉例說明顯現於多結晶陶瓷燒結體的表面 之結晶粒的長度測定之雷射顯微鏡照片。 圖9、圖1 0中的數値是表示測定處及測定號碼。 並且，表5是顯示圖9的測定結果的表，表6是顯示圖 10的測定結果的表。 -25- 201212159 【5揪〕 AVE. 1.509 1.666 CM 2.657 <N <N 0.889 0.789 S 0.656 2 2.835 00 3.145 r- 0.698 v〇 0.699 0.554 2 1.252 m 0.709 rj 0.875 1.498 O 0.631 σ\ 1.334 00 1.643 卜 1.233 rn w-> 1.563 寸 1,825 m 1.854 (N 2.684 3.187 測定 號碼 長度 (um) -26- 201212159 ¥ AVE. 1.629 0.832 m cs 1.032 (N <N 2.404 2.88 1.047 〇\ oo 1.633 二 2.186 v〇 0.898 »〇 0.563 0.708 CO 0.656 <N 2.407 1.374 o »—Η 〇\ 1.249 oo 0.647 卜 0.437 VO 1.201 yn 1.804 寸 1.016 m 0.848 CN 6.106 4.66 測定 號碼 長度 (μπ〇 -27- 201212159 由表5、表6可知，微細的凹部的長度與多結晶陶瓷燒 結體的表面所顯現的結晶粒的長度可謂同程度。 這表示微細的凹部是對應於多結晶陶瓷燒結體的表面 所顯現的結晶粒來形成。 若根據本實施形態的CMP法，則可容易且確實地形成 突起部3a、平坦部3b2、凹部13a、凹部13b。並且，後述 的千涉條紋佔有面積率亦可形成未滿1 %。 (對缺陷部的定量評價法） 其次，說明有關對於介電質基板3的表面區域內部存 在的龜裂等的缺陷部之定量評價法。 首先，說明有關介電質基板3的表面區域內部存在的 龜裂等的缺陷部。 圖Π是用以舉例說明在介電質基板3的表面區域所發 生的龜裂的掃描型電子顯微鏡照片。 圖1 2是用以舉例說明表面區域的一部分似脫離的狀態 的掃描型電子顯微鏡照片。 利用噴沙法等的機械性加工法來形成突起部3a、平面 部3b時，如圖11所示，有時在介電質基板3的表面區域產 生龜裂等的缺陷部。 而且，當如此的缺陷部存在於表面區域內時，如圖1 2 所示，有時表面區域的一部分會似脫離，不久就脫離。 另外，所發生的龜裂，有產生於結晶粒界者、貫通結 晶粒界內而產生者、及該等不規則地連接者等。Further, at this time, the flat portion 3b2 is formed in the opening of the plurality of holes 3b1 which are opened in the flat portion 3b. Further, the front concave portion 13a is formed on the top surface 3a1 of the projection portion 3a, and the fine 13b° is formed in the flat portion 3b2. According to the findings obtained by the present inventors, (Chemical Mechanical Polishing; Chemistry) Mechanical honing) The protrusion 3a, the flat portion 3b2, the recess 13a, and the 1bb can be formed at one time. The CMP method is generally used for planarization processing. Therefore, it is not necessary to form the protrusions 3a having the above-described shapes, and it is not necessary to form the fine recesses 13a and the recesses 13b. However, according to the findings obtained by the inventors, the concave portion 13a and the concave portion 13b can be formed by the action of a chemical component containing a grinding fluid (sluury). In other words, the polycrystalline ceramic sintered body has a recessed portion 13a and a recessed portion 13b which are formed in accordance with the orientation of the etching plane. In the surface region of the polycrystalline ceramic sintered body, the portion which is easily ablated by the crystal plane is first etched, so that fine 13a and recess 13b can be formed. Further, the portion 3a and the flat portion 3b2 can be formed by mechanical honing of the granules contained in the honing liquid and chemical honing effects of the chemical components contained in the honing liquid. In this case, the flat portion 3b2 is formed in? Surroundings. Here, the process conditions of the CMP method are exemplified. The CMP method of the micro-recessed part is described in the periphery, and the recessed portion can be said to have no effect on the knot, and the effect of the concave portion is immediately formed. 3b 1 -23- 201212159 Honing cloth, for example, can be a rigid foamed polyurethane honing cloth, etc. The rotation speed can be set to 60 rpm, and the load can be set to 0.2 kg/cm 2 or the like. The cerium particles contained in the honing liquid may be composed of SiO 2 (cerium oxide), Ce02 (cerium oxide), TiO 2 (titanium oxide), MgO (magnesium oxide), Y 2 〇 3 (cerium oxide), Sn02 (tin oxide), and the like. . Further, the ratio of the granules to the honing liquid can be set to about 10 to 20% by weight. The chemical component contained in the honing liquid may be a pH adjuster, a dispersant of cerium particles, a surfactant, or the like. In this case, in consideration of the above-described crystal anisotropic etching, it is preferable that the honing liquid is alkaline. Therefore, the hydrogen ion index of the honing liquid is about 8 to 13 degrees. Further, the supply amount of the honing liquid can be, for example, about 20 cc / min. Moreover, according to the insights obtained by the inventors, processing time is an important factor. In other words, when the processing time is short, the flattening process is performed, and the projection 3a having the above-described shape cannot be formed, and the fine recessed portion 13a and the recessed portion 13b cannot be formed. For example, a processing time of a few minutes is a flattening process. On the other hand, if the processing time is several hours, the projections 3a having the above-described shapes can be formed, and the fine recesses 1 3 a and the recesses 1 3 b can be formed. Further, since the top surface 3a1 of the projection portion 3a is easier to machine than the flat portion 3b, the relationship between the depth dimension of the recess portion 13a and the depth dimension of the recess portion 13b can be formed. That is, the depth dimension of the concave portion 13a formed in the top surface 3a1 may be larger than the depth dimension of the concave portion 13b formed in the flat portion 3b2. Further, the above processing time can be appropriately changed according to other process conditions of the CMP method, 24-201212159 (for example, the hydrogen ion index of the honing liquid, etc.), and the interference fringe area ratio to be described later can also be considered. In other words, not only the formation of the concave portion 13a or the concave portion 13b but also the processing of the CMP method can be performed until the interference fringe area ratio to be described later forms less than 1%. In addition, the details of the area of the interference fringe occupancy area, etc. will be described later. Further, the height of the projection 3a can be larger than the average particle diameter of the crystal grains of the polycrystalline ceramic sintered body to be described later. Alternatively, the average particle diameter of the crystal grains of the polycrystalline ceramic sintered body may be smaller than the height dimension of the protrusions 3a. When formed in this way, the crystal grains can be suppressed from being threshed from the dielectric substrate 3. Further, even if the crystal grains are threshed, the shape change of the projections 3a can be suppressed. Fig. 9 is a photograph of a laser microscope for illustrating the measurement of the length of a fine recess. Fig. 1 is a laser microscope photograph for illustrating the measurement of the length of crystal grains appearing on the surface of a polycrystalline ceramic sintered body. The number 値 in Fig. 9 and Fig. 10 indicates the measurement site and the measurement number. Further, Table 5 is a table showing the measurement results of Fig. 9, and Table 6 is a table showing the measurement results of Fig. 10. -25- 201212159 [5揪] AVE. 1.509 1.666 CM 2.657 <N <N 0.889 0.789 S 0.656 2 2.835 00 3.145 r- 0.698 v〇0.699 0.554 2 1.252 m 0.709 rj 0.875 1.498 O 0.631 σ\ 1.334 00 1.643 1.233 rn w-> 1.563 inch 1,825 m 1.854 (N 2.684 3.187 determination number length (um) -26- 201212159 ¥ AVE. 1.629 0.832 m cs 1.032 (N <N 2.404 2.88 1.047 〇\ oo 1.633 two 2.186 v〇0.898 »〇0.563 0.708 CO 0.656 <N 2.407 1.374 o »—Η 〇\ 1.249 oo 0.647 Bu 0.437 VO 1.201 yn 1.804 inch 1.016 m 0.848 CN 6.106 4.66 Determination of the length of the number (μπ〇-27- 201212159 is shown in Table 5 and Table 6 The length of the fine concave portion is the same as the length of the crystal grain appearing on the surface of the polycrystalline ceramic sintered body. This means that the fine concave portion is formed corresponding to the crystal grains appearing on the surface of the polycrystalline ceramic sintered body. In the CMP method of the present embodiment, the protrusion 3a, the flat portion 3b2, the recess 13a, and the recess 13b can be easily and surely formed. The area ratio of the area occupied by the stripe may be less than 1%. (Quantitative evaluation method for the defect portion) Next, a quantitative evaluation method for the defect portion such as cracks existing inside the surface region of the dielectric substrate 3 will be described. A defect portion such as a crack existing in the surface region of the dielectric substrate 3 will be described. Fig. Π is a scanning electron micrograph for illustrating cracks occurring in the surface region of the dielectric substrate 3. 2 is a scanning electron micrograph for illustrating a state in which a part of the surface region is detached. When the projection 3a and the flat portion 3b are formed by a mechanical processing method such as a sandblasting method, as shown in FIG. A defect portion such as a crack is generated in the surface region of the dielectric substrate 3. Further, when such a defect portion exists in the surface region, as shown in Fig. 12, a part of the surface region may appear to be detached, and soon Further, the occurrence of cracks may occur in the crystal grain boundary, in the crystal grain boundary, and in the irregular connection.

-28- 201212159 由於這樣脫離的表面區域的一部分會成爲微粒，因此 最好至預定的比例爲止除去缺陷部。因應於此，需要定量 評價缺陷部的發生位置，發生的程度（發生比例）等。 可是，在介電質基板3的表面區域內部存在的龜裂等 的缺陷部是無法從外部直接辨識。亦即，以往對缺陷部之 非破壞的定量評價困難。 其次，說明有關本實施形態對缺陷部的定量評價法。 根據本發明者們所取得的見解，若藉由雷射顯微鏡來 攝取介電質基板3的表面，則會在內部存在缺陷部的部分 產生干涉條紋。亦即，根據來自介電質基板3的表面及缺 陷部的面的2個界面的反射光的光路長差而產生干涉條紋 〇 圖13是用以舉例說明在突起部3a的頂面3al內部存在 缺陷部的情況的雷射顯微鏡照片。另外，圖1 3 ( a )是用 以舉例說明在內部存在缺陷部的部分產生的干涉條紋的雷 射顯微鏡照片’圖13 ( b )是圖13 ( a )的B-B線剖面的掃· 描型電子顯微鏡（SEM ; Scanning Electron Microscope) 照片。又，圖1 3 ( c )是圖1 3 ( b )的D部的擴大照片，圖 13(d)是與圖13(a)同部分的掃描型電子顯微鏡照片》 圖14是用以舉例說明在平面部3b的平坦部3b2內部存 在缺陷部的情況的雷射顯微鏡照片。另外，圖1 4 ( a )是 用以舉例說明在內部存在缺陷部的部分產生的干涉條紋的 雷射顯微鏡照片’圖1 4 ( b )是圖1 4 ( a )的C-C線剖面的 掃描型電子顯微鏡照片。 -29- 201212159 此情況，如圖1 3 ( d )所示’利用掃描型電子顯微鏡 的觀察是無法特定表面區域內部存在的缺陷部。 若根據本實施形態的定量評價法，則如圖1 3 ( a )〜 (c )、圖1 4 ( a ) 、 （ b )所示，可藉由干涉條紋來特定 無法直接從外部來辨識的龜裂等的缺陷部。這意味可在非 破壞下定量評價缺陷部的發生或發生的程度等。 並且，亦可根據干涉條紋的大小、方向、週期數等來 得知缺陷部的狀態。 而且，利用如此的干涉條紋的定量評價，可在生產線 中按各靜電吸盤進行。因此，可謀求靜電吸盤的品質、可 靠度、生產性的提升等。 其次，更說明有關利用干涉條紋之對缺陷部的定量評 價。首先，利用雷射顯微鏡來攝取干涉條紋。 雷射顯微鏡是可利用以下者。 掃描型共焦點雷射顯微鏡（〇1>^^118〇〇10〇^1丨〇11〇1^-110 0) 雷射種 ：Ar 波長 ：4 8 8 n m 攝影透鏡：x50對物透鏡zooml 光學模式：非共焦點 雷射強度：1 0 0 檢測感度：442 offset : -1 6 攝影像 ：亮度像-28- 201212159 Since a part of the surface area thus separated becomes fine particles, it is preferable to remove the defective portion up to a predetermined ratio. In response to this, it is necessary to quantitatively evaluate the occurrence position of the defective portion, the degree of occurrence (proportion of occurrence), and the like. However, the defective portion such as cracks existing inside the surface region of the dielectric substrate 3 cannot be directly recognized from the outside. That is, it has been difficult to quantitatively evaluate the non-destructiveness of the defective portion in the past. Next, a quantitative evaluation method for the defective portion according to the present embodiment will be described. According to the findings obtained by the inventors of the present invention, when the surface of the dielectric substrate 3 is taken up by a laser microscope, interference fringes are generated in the portion where the defect portion is present. That is, interference fringes are generated based on the optical path length difference of the reflected light from the two interfaces of the surface of the dielectric substrate 3 and the surface of the defective portion. FIG. 13 is for exemplifying the presence of the top surface 3al of the protruding portion 3a. Laser microscope photograph of the case of the defective part. In addition, FIG. 13( a ) is a laser micrograph for illustrating an interference fringe generated in a portion where a defect portion is present. FIG. 13( b ) is a scanning type of the BB line cross section of FIG. 13 ( a ). Electron Microscope (SEM; Scanning Electron Microscope) photo. Further, Fig. 13 (c) is an enlarged photograph of the D portion of Fig. 13 (b), and Fig. 13 (d) is a scanning electron micrograph of the same portion as Fig. 13 (a). Fig. 14 is a view for exemplification A laser microscope photograph in the case where a defective portion exists inside the flat portion 3b2 of the flat portion 3b. In addition, Fig. 14 (a) is a laser micrograph for illustrating an interference fringe generated in a portion where a defect portion is present. Fig. 14 (b) is a scanning type of a CC line cross section of Fig. 14 (a). Electron micrograph. -29- 201212159 In this case, as shown in Fig. 13 (d), observation by a scanning electron microscope is such that a defect portion existing inside a specific surface region cannot be specified. According to the quantitative evaluation method of the present embodiment, as shown in FIGS. 13( a ) to (c ) and FIGS. 14 ( a ) and ( b ), interference fringes can be used to specify that the identification cannot be directly recognized from the outside. Defective parts such as cracks. This means that the degree of occurrence or occurrence of the defective portion can be quantitatively evaluated without damage. Further, the state of the defective portion can be known from the size, direction, number of cycles, and the like of the interference fringe. Moreover, the quantitative evaluation of such interference fringes can be carried out in the production line for each electrostatic chuck. Therefore, it is possible to improve the quality, reliability, and productivity of the electrostatic chuck. Secondly, the quantitative evaluation of the defective portion using the interference fringes will be described. First, a laser microscope is used to take up interference fringes. Laser microscopes are available for use. Scanning confocal laser microscope (〇1>^^118〇〇10〇^1丨〇11〇1^-110 0) Laser species: Ar Wavelength: 4 8 8 nm Photographic lens: x50 optic lens zooml optics Mode: Non-confocal laser intensity: 1 0 0 Detection sensitivity: 442 offset : -1 6 Photo: brightness image

-30- 201212159 攝影 ：快照累計8張 首先，將介電質基板3或設於靜電吸盤1的介電質基板 3載置於雷射顯微鏡的平台。 然後，將所欲計測的區域（所欲攝影的區域）移動至 對物透鏡的正下方。其次’選擇對物透鏡的倍率等來決定 攝影視野。 並且，以「非共焦點模式」的快照（累計8張）來攝 影。因爲若爲「共焦點模式」，則會產生亮度不均，畫像 處理計測時，用以抽出干涉條紋的臨界値設定變難。另外 ，「非共焦點模式」亦可取得充分的分解能。 其次，將利用雷射顯微鏡所攝取的畫像予以畫像處理 (2値化處理）計測。 圖15是用以舉例說明被2値化處理的畫像的照片。 另外，照片中的亮點狀部分E爲具有干涉條紋的部分 〇 畫像處理計測可利用以下的畫像處理軟體來進行。 畫像處理軟體：Win-ROOF (三谷商事） 2値化處理 ：2800-4095 畫像處理 ：削除〇.2μιη>、塡補空白 測定 ：面積率 其次，根據畫像處理計測的結果來進行對缺陷部的定 量評價。對缺陷部的定量評價，可根據干涉條紋佔有面積 率（對畫像面積之干涉條紋部分面積的比）來進行。例如 圖15的情況，干涉條紋佔有面積率是〇.9 7 %程度。 -31 - 201212159 根據本發明者們所取得的見解，只要利用雷射顯微鏡 所求取之載置被吸附物側的主面的干涉條紋佔有面積率未 滿1%，便可使表面區域的一部分脫離而發生的微粒數大幅 度地減少。 此情況，在介電質基板3的表面區域內部存在的缺陷 部是拋光硏磨法無法除去。並且，利用砥石加工法、雷射 雕刻法、噴砂處理法等，恐有缺陷部更增加之虞。 因此，在本實施形態中是利用前述的CMP法來形成突 起部3a、平坦部3b2、凹部13a、凹部13b，且至干涉條紋 佔有面積率未滿1 %爲止進行內部存在的缺陷部的除去。 圖16是用以舉例說明利用CM P法來除去缺陷部的狀態 的圖表。 圖17是用以舉例說明利用CMP法來除去缺陷部之前的 狀態的圖表。 另外，圖16、圖17是使用於利用庫倫力的靜電吸盤的 介電質基板3的情況。 使用於利用庫倫力的靜電吸盤之介電質基板3，例如 可舉由多結晶陶瓷燒結體所形成，氧化鋁含有率爲99.9wt% 以上，體積密度爲3.96以上，體積電阻率（volume resistivity )在靜電吸盤的使用溫度區域中爲10ΜΩcm以上 者。 另外，本說明書的體積密度是藉由JIS規格（JISR1634 )所示的阿基米德法來測定的値。此情況，飽水方法可爲 真空法，溶媒可使用蒸餾水》 -32- 201212159 如圖1 7所示，即使干涉條紋佔有面積率爲最大3.5 %程 度的狀態時，還是可藉由使用本實施形態的CMP法，如圖 1 6所示，形成干涉條紋佔有面積率爲未滿1 %的狀態。 圖1 8是用以舉例說明利用CMP法來除去缺陷部的狀態 的圖表。 圖19是用以舉例說明利用CMP法來除去缺陷部之前的 狀態的圖表。 另外，圖18、圖19是使用於利用強生·羅貝克力的靜 電吸盤的介電質基板3的情況。 使用於利用強生·羅貝克力的靜電吸盤之介電質基板 3，例如可舉由多結晶陶瓷燒結體所形成，氧化鋁含有率 爲99.4wt%以上，體積電阻率在靜電吸盤的使用溫度區域 中爲108Qcm以上、1013Qcm以下者。 如圖1 9所示，即使干涉條紋佔有面積率爲最大5%程度 的狀態時，還是可藉由使用本實施形態的CMP法，如圖1 8 所示，形成干涉條紋估有面積率爲未滿1 %的狀態。 亦即，即使介電質基板3的組成改變，還是可藉由使 用前述的CMP法，形成干涉條紋佔有面積率爲未滿1%的狀 態。 在此，介電質基板3的體積電阻率可在燒結時控制。 其次’說明介電質基板3的製造方法。 首先，準備氧化鋁及氧化鈦作爲原料。所使用的氧化 鋁及氧化鈦較理想是微粒者，氧化鋁粉末是平均粒子徑 0.3μηι以下’更理想是使用〇.2μιη以下者。另—方面，氧化 -33- 201212159 鈦粉末是平均粒子徑Ο.ίμιη以下，更理想是使用0.05μιη以 下者。在原料使用微粒粉之下，分散佳，不易形成粒子徑 大的鈦化合物的偏析物。 另外，較理想的氧化鋁粉末的平均粒子徑的下限値是 1 Onni。又，較理想的氧化鈦粉末的平均粒子徑的下限値是 5nm。 其次，進行泥漿調整、造粒、生胚加工。 將上述原料予以預定量秤量，更加上分散劑·黏合劑 •離型劑，進行利用球磨機的粉碎攪拌混合。在混合中使 用離子交換水等，不使雜質混入爲理想。混合後進行利用 噴霧乾燥機的造粒，在將所取得的造粒粉末沖壓成形下， 可製作生胚成形體。而且，上述生胚成形體是進行CIP成 形爲理想。在CIP成形下，生胚成形體的密度會上升，可 提高燒結體的密度。另外，成形並非限於乾式成形，利用 擠壓成形、射出成形、薄板成形、熔鑄成形、凝膠注模成 形等的成形方法，也可取得生胚成形體。 其次，進行燒結》 在氮、氫氣還原環境下燒結上述生胚成形體，藉此可 製造介電質基板3。 在進行還原燒結下，氧化鈦會成爲非化學計量組成， 可控制體積電阻率。 例如’在進行以下那樣的燒結下，可製造體積電阻率 在靜電吸盤的使用溫度區域中爲1〇8Ωοιη以上、1013Qcm以 下的介電質基板3。 -34- 201212159 燒結溫度可爲1150〜1 350°c，更理想是1 150〜1 200°C 的溫度範圍。藉由在低溫下燒結，可抑.制氧化鋁粒子的粒 子成長，亦可抑制偏析的鈦化合物的成長。因此，可更縮 小氧化鋁粒子的最大粒子徑。 並且，在燒結的最高溫度的保持時間，爲了使燒結體 的物性値安定化，最好是2小時以上，更理想是4小時以上 〇 另外，所取得的燒結體是更施以Η IP處理爲理想。藉 此可取得緻密質的介電質基板3。 如以上那樣，可製造介電質基板3。 若根據本實施形態對缺陷部的定量評價法，則可在非 破壞下定量評價缺陷部的發生或發生的程度等。並且，可 根據此定量評價來形成干涉條紋佔有面積率爲未滿1 %。而 且，此定量評價可在生產線中按各靜電吸盤進行。因此， 可使在表面區域的一部分脫離下產生的微粒數大幅度地減 少。而且，可謀求靜電吸盤的品質、可靠度、生產性的提 升等。 另外，舉例說明定量評價在介電質基板3的表面區域 內部存在的龜裂等缺陷部的情況，但亦可定量評價其他的 形態在介電質基板的表面區域內部存在的龜裂等的缺陷部 。例如，亦可定量評價形成有未被形成凹部13a、凹部13b 的突起部及平面部的介電質基板、或未形成有突起部及平 面部的平板狀的介電質基板等的表面區域內部存在的龜裂 等的缺陷部。' -35- 201212159 (多結晶陶瓷燒結體的結晶粒的平均粒子徑） 其次，說明有關構成介電質基板3的多結晶陶瓷燒結 體的結晶粒的平均粒子徑。 首先，說明有關結晶粒的平均粒子徑的測定。 將成爲測定對象的多結晶陶瓷燒結體的表面加工成無 傷的鏡面。鏡面加工可藉由鑽石硏磨法來進行。然後，熱 蝕刻被鏡面加工的面。熱蝕刻的條件是溫度可爲1 3 3 0 °C程 度，時間爲2小時程度。 其次’在表面上濺射塗層Au (金）。塗層的厚度可爲 2〇nm程度。Au (金）的濺射塗層的目的是在於使用雷射 顯微鏡時，可使結晶粒界的對比鮮明。亦即，Au (金）的 濺射塗層是爲了防止雷射光侵入至多結晶陶瓷燒結體的內 部而實施者。-30- 201212159 Photograph: 8 snapshots are accumulated First, the dielectric substrate 3 or the dielectric substrate 3 provided on the electrostatic chuck 1 is placed on a platform of a laser microscope. Then, the area to be measured (the area to be photographed) is moved directly below the objective lens. Next, 'the magnification of the objective lens is selected to determine the photographic field of view. Also, take a snapshot of the "non-confocal mode" (accumulated 8 shots). If the "co-focus mode" is used, uneven brightness will occur, and the threshold setting for extracting interference fringes will be difficult to achieve during image processing. In addition, the "non-confocal mode" can also achieve sufficient decomposition energy. Next, the image taken by the laser microscope is subjected to image processing (2 値 processing) measurement. Fig. 15 is a photograph for illustrating an example of a double-twisted image. Further, the bright spot portion E in the photograph is a portion having interference fringes. The image processing measurement can be performed by the following image processing software. Image processing software: Win-ROOF (Sangu business) 2 値 processing: 2800-4095 Image processing: 〇 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定 测定Evaluation. The quantitative evaluation of the defective portion can be performed based on the area ratio of the interference fringe (the ratio of the area of the interference fringe portion of the image area). For example, in the case of Fig. 15, the interference fringe occupancy area ratio is about 9.7%. -31 - 201212159 According to the findings obtained by the present inventors, a portion of the surface region can be obtained by occupying an area ratio of interference fringes of the main surface on the side of the object to be adsorbed which is obtained by a laser microscope to less than 1%. The number of particles that occur when detached is greatly reduced. In this case, the defective portion existing inside the surface region of the dielectric substrate 3 cannot be removed by the polishing honing method. In addition, the use of vermiculite processing, laser engraving, sandblasting, etc., may increase the number of defective parts. Therefore, in the present embodiment, the protruding portion 3a, the flat portion 3b2, the concave portion 13a, and the concave portion 13b are formed by the above-described CMP method, and the defect portion existing inside is removed until the interference fringe occupying area ratio is less than 1%. Fig. 16 is a graph for explaining a state in which a defective portion is removed by the CM P method. Fig. 17 is a graph for explaining a state before the defective portion is removed by the CMP method. Further, Fig. 16 and Fig. 17 show a case where the dielectric substrate 3 is used for an electrostatic chuck using Coulomb force. The dielectric substrate 3 used for the electrostatic chuck using Coulomb force is, for example, formed of a polycrystalline ceramic sintered body, and has an alumina content of 99.9 wt% or more, a bulk density of 3.96 or more, and a volume resistivity. It is 10 Μ Ωcm or more in the use temperature region of the electrostatic chuck. In addition, the bulk density of this specification is the enthalpy measured by the Archimedes method shown by JIS specification (JISR1634). In this case, the saturated water method may be a vacuum method, and the solvent may be distilled water. -32 - 201212159 As shown in Fig. 17, even if the interference fringe occupancy area is at a maximum of 3.5%, the embodiment can be used. The CMP method, as shown in Fig. 16, forms a state in which the interference fringe occupancy area ratio is less than 1%. Fig. 18 is a graph for explaining a state in which the defective portion is removed by the CMP method. Fig. 19 is a graph for explaining a state before the defective portion is removed by the CMP method. Further, Fig. 18 and Fig. 19 show a case where the dielectric substrate 3 is used for an electrostatic chuck using a Johnson & Johnson - Robin force. The dielectric substrate 3 used for the electrostatic chuck using the Johnson & Loebeck force is, for example, formed of a polycrystalline ceramic sintered body, and has an alumina content of 99.4% by weight or more and a volume resistivity in the use temperature region of the electrostatic chuck. The middle is 108Qcm or more and 1013Qcm or less. As shown in Fig. 19, even if the interference fringe occupies a region where the area ratio is at most 5%, the CMP method of the present embodiment can be used, as shown in Fig. 18, the area ratio of the interference fringes is estimated to be not Full 1% status. That is, even if the composition of the dielectric substrate 3 is changed, the interference fringe occupying area ratio of less than 1% can be formed by using the CMP method described above. Here, the volume resistivity of the dielectric substrate 3 can be controlled at the time of sintering. Next, a method of manufacturing the dielectric substrate 3 will be described. First, alumina and titanium oxide are prepared as raw materials. The aluminum oxide and the titanium oxide to be used are preferably fine particles, and the alumina powder has an average particle diameter of 0.3 μη or less. More preferably, it is 0.25 μm or less. On the other hand, oxidation -33 - 201212159 Titanium powder is an average particle diameter Ο. ίμιη or less, more preferably 0.05 μmη or less. Under the use of the fine particles of the raw material, the dispersion is good, and it is difficult to form segregation of the titanium compound having a large particle diameter. Further, the lower limit 平均 of the average particle diameter of the preferred alumina powder is 1 Onni. Further, the lower limit 平均 of the average particle diameter of the titanium oxide powder which is more preferable is 5 nm. Secondly, mud adjustment, granulation, and green embryo processing are carried out. The raw materials are weighed in a predetermined amount, and the dispersant, the binder, and the release agent are further added, and the mixture is stirred and mixed by a ball mill. It is preferable to use ion-exchanged water or the like in mixing without mixing impurities. After mixing, granulation by a spray dryer is carried out, and a green body molded body can be produced by press-forming the obtained granulated powder. Further, the green preform is preferably formed by CIP molding. Under the CIP molding, the density of the green body formed body increases, and the density of the sintered body can be increased. Further, the molding is not limited to dry molding, and a green body molded body can be obtained by a molding method such as extrusion molding, injection molding, sheet molding, melt casting molding, or gel injection molding. Next, sintering is carried out to sinter the green preform in a nitrogen or hydrogen reducing atmosphere, whereby the dielectric substrate 3 can be produced. Under reduction sintering, titanium oxide becomes a non-stoichiometric composition that controls volume resistivity. For example, under the following sintering, the dielectric substrate 3 having a volume resistivity of 1 〇 8 Ω οηη or more and 1013 Q cm or less in the use temperature region of the electrostatic chuck can be manufactured. -34- 201212159 Sintering temperature can be 1150~1 350 °c, more ideally 1 150~1 200 °C temperature range. By sintering at a low temperature, the growth of the particles of the alumina particles can be suppressed, and the growth of the segregated titanium compound can be suppressed. Therefore, the maximum particle diameter of the alumina particles can be further reduced. Further, in order to stabilize the physical properties of the sintered body in the holding time of the highest temperature of sintering, it is preferably 2 hours or longer, more preferably 4 hours or longer, and the obtained sintered body is further subjected to Η IP treatment. ideal. Thereby, the dense dielectric substrate 3 can be obtained. As described above, the dielectric substrate 3 can be manufactured. According to the quantitative evaluation method of the defective portion according to the present embodiment, the degree of occurrence or occurrence of the defective portion can be quantitatively evaluated without damage. Further, the interference fringe occupancy area ratio can be formed to be less than 1% based on the quantitative evaluation. Moreover, this quantitative evaluation can be carried out in the production line for each electrostatic chuck. Therefore, the number of particles generated by detaching a part of the surface region can be greatly reduced. Further, it is possible to improve the quality, reliability, and productivity of the electrostatic chuck. Further, a case where a defect such as a crack existing in the surface region of the dielectric substrate 3 is quantitatively evaluated is exemplified, but a defect such as a crack existing in the surface region of the dielectric substrate in another form may be quantitatively evaluated. unit. For example, it is also possible to quantitatively evaluate the surface area of a dielectric substrate on which a projection portion and a flat portion in which the concave portion 13a and the recess portion 13b are not formed, or a flat dielectric substrate on which the projection portion and the flat portion are not formed. Defects such as cracks. '-35-201212159 (Average particle diameter of crystal grains of the polycrystalline ceramic sintered body) Next, the average particle diameter of the crystal grains of the polycrystalline ceramic sintered body constituting the dielectric substrate 3 will be described. First, the measurement of the average particle diameter of the crystal grains will be described. The surface of the polycrystalline ceramic sintered body to be measured is processed into an innocuous mirror surface. Mirror processing can be performed by diamond honing. Then, the mirror-finished surface is thermally etched. The condition of the thermal etching is that the temperature can be in the range of 1 3 30 ° C for a period of 2 hours. Next, a coating of Au (gold) was sputtered on the surface. The thickness of the coating can be about 2 〇 nm. The purpose of the Au (gold) sputter coating is to make the contrast of the crystal grain boundaries clear when using a laser microscope. That is, the sputtering coating of Au (gold) is carried out in order to prevent laser light from intruding into the interior of the polycrystalline ceramic sintered body.

Au (金）的濺射塗層可利用離子濺射裝置（日立製作 所製、E-105 )等來進行。 其次’將熱蝕刻後的多結晶陶瓷燒結體，利用雷射顯 微鏡來攝取多結晶陶瓷燒結體。 將多結晶陶瓷燒結體載置於雷射顯微鏡的平台。然後 ’將所欲計測的區域（所欲攝影的區域）移動於對物透鏡 的正下方。其次’選擇對物透鏡的倍率等來決定攝影視野 〇 並且’以「非共焦點模式」的快照（累計8張）來攝 影。因爲若爲「共焦點模式」，則會雷射光產生亮度不均 -36- 201212159 ，畫像處理計測時’用以抽出結晶粒界的臨界値設定變難 。另外，「非共焦點模式」亦可取得充分的分解能。 雷射顯微鏡可使用以下者。 掃描型共焦點雷射顯微鏡（01ymPus C〇rPoration 0LS_ 1100) 雷射種類：ΑΓ 波長 ：4 8 8 n m 攝影透鏡：xlOO對物透鏡zooml 光學模式：非共焦點 雷射強度：1 〇 〇 檢測感度：400 offset : -30 攝影像 ：亮度像 攝影 ：快照累計8張 圖20是用以舉例說明藉由雷射顯微鏡來攝取的多結晶 陶瓷燒結體的照片。另外，圖20是使用於利用庫倫力的靜 電吸盤的介電質基板3的情況。又，圖20(a)是結晶粒的 平均粒子徑爲1_8μηι程度的情況，圖20 ( b )是結晶粒的平 均粒子徑爲1.4 μπι程度的情況。 圖2 1是用以舉例說明藉由雷射顯微鏡來攝取的多結晶 陶瓷燒結體的照片。另外，圖2 1是使用於利用強生•羅貝 克力的靜電吸盤的介電質基板3的情況。 又’圖2 1是結晶粒的平均粒子徑爲1 μιη程度的情況^ 其次’根據利用雷射顯微鏡所攝取的畫像來求取多結 -37- 201212159 晶陶瓷燒結體的結晶粒的平均粒子徑。 結晶粒的平均粒子徑的運算可利用以下的軟體來進行 0 畫像處理軟體 ：Win-ROOF (三谷商事） 校準（Calibration ) : 0 · 1 25 μηι/ρixe 1 背景處理 ：12.5gm/100pixcel 2値化處理 ：2100-2921 圓形分離 ：自動處理 計測 ：圓當量直徑 圖22是用以舉例說明結晶粒的平均粒子徑與粒子徑分 布的標準偏差的圖表。 圖23也是用以舉例說明結晶粒的平均粒子徑與粒子徑 分布的標準偏差的圖表。 另外，圖22是使用於利用庫倫力的靜電吸盤的介電質 基板3的情況，圖23是使用於利用強生•羅貝克力的靜電 吸盤的介電質基板3的情況。 若根據本發明者所取得的見解，則只要將結晶粒的平 均粒子徑設爲〇.8μηι以上、1.5μιη以下，便可抑制結晶粒從 介電質基板3的表面脫粒。其結果，可抑制微粒的發生。 並且，即使假設發生脫粒，還是會因爲粒子徑小，所以難 被保持於凹凸部，可容易除去。而且，可抑制因脫粒而突 起部3a等的形狀變化。 又，若將粒子徑分布的標準偏差設爲1 μηι以下，則可 更抑制結晶粒從介電質基板3的表面脫粒。並且，即使假 -38- 201212159 設結晶粒脫粒，還是可抑制突起部3 a的形狀變化。 此情況，可藉由控制燒結條件來控制結晶粒的平均粒 子徑的範圍。例如，藉由控制燒結溫度（例如1 3 70°C程度 )或溫度履歷等來阻礙結晶粒的成長即可。 (微細的凹部的深度尺寸） 其次，說明有關微細的凹部的測定。 雷射顯微鏡可使用以下者。 掃描型共焦點雷射顯微鏡（Olympus Corporation OLS-1100) 攝影條件可爲以下般。 雷射種類：Ar 波長 ：48 8nm 攝影透鏡：xlOO對物透鏡zo〇m4.0 光學模式：共焦點 雷射強度：1 0 0 檢測感度：4 0 0 offset : 0 畫像取入模式：3次元取入（上下限） step量 ：0.0 1 μιη 攝影像 ：亮度像 攝影 ：快照累計8張 攝影可按以下的程序進行。 首先’將設於介電質基板3或靜電吸盤1的介電質基板 -39- 201212159 3載置於雷射顯微鏡的平台。 將所欲測定的區域（所欲攝影的區域）移動至對物透 鏡的正下方。 其次，選擇對物透鏡的倍率等，而形成攝影倍率。 將光學模式設定成共焦點，設定高度方向的取入條件 ，而進行畫像攝影。 微細的凹部的深度測定條件是可設爲以下般。 測定模式：段差測定 剖面方向：水平及垂直 平均模式 ：線 剖面寬度 ：1 點（point ) :波形位置 圖24是用以舉例說明微細的凹部的深度測定。另外， 圖24 ( a )是用以舉例說明測定値的分布的圖表，圖24 ( b )是用以舉例說明測定位置的雷射顯微鏡照片.。 微細的凹部的深度測定是可按以下的程序進行。 首先，在攝取的畫像中，設定測定條件。 如在圖24 ( a ) 、 ( b )所例示般，逐一比對水平方向 及垂直方向的測定値的分布，計測1 2點以上畫像內的凹凸 大之處。但，缺陷部1 00 (因脫粒而形成者）除外。 測定的1 2點以上的段差之中，將最大的段差設爲凹凸 Μ AX。在此，形成於頂面3 a 1的微細的凹部1 3 a是按以下的 程序求取。從介電質基板3或靜電吸盤1的中心朝外周三等 配間距（測定畫像爲4張）以上進行。 -40- 201212159 將各測定位置的凹凸段差的MAX値之中最大値設爲形 成於頂面3al的微細的凹部13 a的深度尺寸。 圖25是表示形成於頂面3al的微細的凹部13 a的深度尺 寸與附著於被吸附物的背面的微粒數的關係圖表。 根據前述的攝影條件、測定條件來測定樣品1〜3的凹 部1 3a的深度尺寸。 樣品1的凹部13a的深度尺寸是l5〇nm程度，樣品2的凹 部13a的深度尺寸是30nrn程度，樣品3的凹部13a的深度尺 寸是20nm程度。 若凹部13a的深度尺寸形成20ηιη程度，則附著於被吸 附物的背面的微粒數是形成6 00個。 相對的，只要將凹部13a的深度尺寸設爲30nm以上、 15 Onm以下，便可使附著於被吸附物的背面的微粒數成爲 2 5 0個以下。 並且，若凹部13 a的深度尺寸超過15 Onm，則進入凹部 13 a的內部之微粒的除去困難。 因此，微細的凹部13a的深度尺寸是30nm以上、150nm 以下爲理想。 (多結晶氧化鋁燒結體的體積密度與氧化鋁含有率） 爲了利用前述的CMP法來形成微細的凹部13a或凹部 13b，成爲底層的多結晶陶瓷燒結體的體積密度及純度（ 含有率）重要。 在此’舉一例說明多結晶氧化鋁燒結體的情況。 -41 - 201212159 圖2 6是多結晶氧化鋁燒結體表面的掃描型電子顯微鏡 照片。另外’圖26 (a)是結晶粒的平均粒子徑爲20μηι〜 50μηι，體積密度爲3.7，氧化鋁含有率爲90vvt%的情況。圖 26 ( b )是結晶粒的平均粒子徑爲1·5μιη以下，體積密度爲 3.96，氧化鋁含有率爲99.9wt %的情況。 由圖26 ( a )與圖26 ( b )的比較可知，若將體積密度 設爲3 · 9 6以上’將氧化鋁含有率設爲9 9 · 9 w t %以上，則可 將成爲底層的多結晶氧化鋁燒結體形成緻密的組織，因此 可更確實地抑制結晶粒從介電質基板3脫粒。 此情況，只要將結晶粒的平均粒子徑設爲0.8 μπι以上 ，1 5 μιη以下’便可成爲緻密的組織。另外，只要將體積 密度及純度（含有率）的至少其中一方設於預定的範圍內 ’便可成爲緻密的組織。但，如前述者般，將體積密度及 純度（含有率）的雙方設於預定的範圍內爲理想。更理想 是將結晶粒的平均粒子徑設爲〇.8μηι以上、1.5μιη以下。此 情況，如前述般，粒子徑分布的標準偏差是形成1 μπι以下 爲理想。 只要將成爲底層的多結晶陶瓷燒結體設爲緻密的組織 ’便可利用前述的CMP法來均一且安定地形成微細的凹部 13 a或凹部13b。其結果’可使微粒的發生大幅度地減少。 此情況’體積密度可藉由進行Η IP處理（熱等靜壓加 工）等來控制。並且’結晶粒的平均粒子徑是如前述般可 依據燒結條件（燒結溫度或溫度履歷等）來控制。 圖27是用以舉例說明附著於半導體晶圓的背面的微粒 -42- 201212159 數的模式圖。另外，圖27 (a)是成爲底層的多結晶氧化 鋁燒結體爲圖26(a)所示者的情況’圖27 (b)是成爲底 層的多結晶氧化鋁燒結體爲圖26(b)所示者的情況》 圖27 (a)的情況是附著於8英吋半導體晶圓的背面的 微粒數爲1058個，圖27(b)的情況是附著於8英吋半導體 晶圓的背面的微粒數爲6 7個。 (靜電吸盤的其他實施形態） 圖28是用以舉例說明其他的實施形態的靜電吸盤la的 模式剖面圖。另外，圖28 (a)是用以舉例說明靜電吸盤 的模式剖面圖，圖28(b)是圖28(a)的F部的模式擴大 圖。 在本實施形態的靜電吸盤la中，是在介電質基板30的 內部埋入電極4。 如此的靜電吸盤1 a是例如利用生胚薄片印刷層疊法等 來製造。 例如，首先，在由多結晶陶瓷燒結體（例如多結晶氧 化鋁燒結體）所構成的生胚薄片，網版印刷鎢膏，藉此形 成電極。然後，以能夠埋設電極的方式，加壓層疊複數的 生胚薄片’形成燒結前的層疊體。將此層疊體切削加工成 所望的形狀，在還原環境中燒結，藉此可製造電極被埋設 於內部的介電質基板30。 (靜電吸盤的製造方法） -43- 201212159 其次，舉例說明有關本實施形態的靜電吸盤的製造方 法。 另外，設於靜電吸盤的介電質基板3是可如前述般製 造。並且，有關電極4等各要素的形成、接合、安裝等的 工程可適用已知的技術，因此該等的說明省略，只說明特 徵性的工程。 圖29是用以舉例說明本實施形態的靜電吸盤的製造方 法的流程圖。 首先，利用已知的噴沙法等，在介電質基板3之載置 被吸附物側的主面形成突起部3a、平面部3b的大部分的形 狀。 其次，如圖29所示，利用前述的CMP法來形成突起部 3a、平坦部3b2、凹部13a、凹部13b。 此時，利用對前述缺陷部的定量評價法來求取缺陷部 所佔的比例，至缺陷部所佔的比例成爲預定的値以下爲止 繼續CMP法的加工。 亦即，至利用雷射顯微鏡所求取的上述主面的干涉條 紋佔有面積率未滿1 %爲止繼續上述主面的加工。 另外，有關CMP法、對缺陷部的定量評價法等的詳細 是可與前述同樣，因此詳細的說明省略。 [產業上的利用可能性] 像以上詳述那樣，若根據本發明，則可提供—種能夠 抑制微粒的發生之靜電吸盤及靜電吸盤的製造方法’產業The sputter coating of Au (gold) can be carried out by using an ion sputtering apparatus (E-105, manufactured by Hitachi, Ltd.). Next, the polycrystalline ceramic sintered body after hot etching is taken up by a laser microscope to take up the polycrystalline ceramic sintered body. The polycrystalline ceramic sintered body is placed on a platform of a laser microscope. Then, 'the area to be measured (the area to be photographed) is moved directly below the objective lens. Next, 'the magnification of the objective lens is selected to determine the photographic field of view 〇 and the snapshot is taken in the "non-focal focus mode" (accumulated 8 frames). In the case of the "co-focus mode", the laser light is unevenly colored -36 - 201212159, and it is difficult to set the threshold of the crystal grain boundary during image processing measurement. In addition, the "non-confocal mode" can also achieve sufficient decomposition energy. The following can be used for the laser microscope. Scanning Confocal Laser Microscope (01ymPus C〇rPoration 0LS_ 1100) Laser Type: ΑΓ Wavelength: 4 8 8 nm Photographic Lens: xlOO Opposite Lens Zooml Optical Mode: Non-Confocal Laser Intensity: 1 〇〇 Detection Sensitivity: 400 offset : -30 Photographic image: Brightness Photographing: 8 snapshots are accumulated. Fig. 20 is a photograph for illustrating a polycrystalline ceramic sintered body taken by a laser microscope. Further, Fig. 20 shows a case where the dielectric substrate 3 is used for an electrostatic chuck using a Coulomb force. Further, Fig. 20(a) shows a case where the average particle diameter of the crystal grains is about 1-8 μm, and Fig. 20(b) shows a case where the average particle diameter of the crystal grains is about 1.4 μm. Fig. 21 is a photograph for illustrating a polycrystalline ceramic sintered body taken up by a laser microscope. Further, Fig. 21 is a case of using the dielectric substrate 3 of an electrostatic chuck using Johnson & Johnson. Further, Fig. 2 is a case where the average particle diameter of the crystal grains is about 1 μm. Next, 'the average particle diameter of the crystal grains of the sintered ceramic body is obtained from the image taken by the laser microscope. 37-201212159 . The calculation of the average particle diameter of the crystal grains can be performed by the following software. 0 Image processing software: Win-ROOF (Calibration): 0 · 1 25 μηι/ρixe 1 Background processing: 12.5gm/100pixcel 2 Treatment: 2100-2921 Circular separation: automatic processing measurement: round equivalent diameter FIG. 22 is a graph for illustrating the standard deviation of the average particle diameter and the particle diameter distribution of the crystal grains. Fig. 23 is also a graph for illustrating the standard deviation of the average particle diameter and particle diameter distribution of crystal grains. Further, Fig. 22 shows a case where the dielectric substrate 3 is used for an electrostatic chuck using Coulomb force, and Fig. 23 shows a case where the dielectric substrate 3 is used for an electrostatic chuck using Johnson & Johnson. According to the findings obtained by the present inventors, it is possible to suppress the degranulation of crystal grains from the surface of the dielectric substrate 3 by setting the average particle diameter of the crystal grains to 〇.8 μηι or more and 1.5 μm or less. As a result, the occurrence of fine particles can be suppressed. Further, even if it is assumed that threshing occurs, the particle diameter is small, so that it is difficult to be held in the uneven portion and can be easily removed. Further, it is possible to suppress the shape change of the protruding portion 3a or the like due to the threshing. Further, when the standard deviation of the particle diameter distribution is 1 μη or less, the crystal grains can be more suppressed from being threshed from the surface of the dielectric substrate 3. Further, even if the crystal grain is threshed, the shape change of the protrusion 3 a can be suppressed. In this case, the range of the average particle diameter of the crystal grains can be controlled by controlling the sintering conditions. For example, it is sufficient to control the growth of the crystal grains by controlling the sintering temperature (for example, about 1 3 to 70 ° C) or the temperature history. (Depth Dimensions of Fine Concave Portions) Next, the measurement of the fine concavities will be described. The following can be used for the laser microscope. Scanning confocal laser microscope (Olympus Corporation OLS-1100) The shooting conditions can be as follows. Laser type: Ar Wavelength: 48 8nm Photographic lens: xlOO object lens zo〇m4.0 Optical mode: Confocal laser intensity: 1 0 0 Detection sensitivity: 4 0 0 offset : 0 Image capture mode: 3 dimensional acquisition In (upper and lower limit) Step amount: 0.0 1 μιη Photo: Brightness photography: Snapshot accumulation of 8 shots can be performed by the following procedure. First, the dielectric substrate -39 - 201212159 3 provided on the dielectric substrate 3 or the electrostatic chuck 1 is placed on the platform of the laser microscope. Move the area to be measured (the area to be photographed) directly below the objective lens. Next, the magnification of the objective lens or the like is selected to form a photographic magnification. The image mode is performed by setting the optical mode to the confocal point and setting the entry condition in the height direction. The depth measurement condition of the fine concave portion can be set as follows. Measurement mode: Step difference measurement Section direction: horizontal and vertical Average mode: Line Section width: 1 point (point): Wave position Position Fig. 24 is a diagram for illustrating the depth measurement of the fine concave portion. In addition, Fig. 24 (a) is a graph for illustrating the measurement of the distribution of ruthenium, and Fig. 24 (b) is a laser microscope photograph for exemplifying the measurement position. The depth measurement of the fine recess can be performed by the following procedure. First, the measurement conditions are set in the image taken. As shown in Fig. 24 (a) and (b), the distribution of the measurement 値 in the horizontal direction and the vertical direction is compared one by one, and the unevenness in the image at 12 or more points is measured. However, the defective portion 100 (formed by degranulation) is excluded. Among the measured step points of 12 points or more, the largest step difference is set as the unevenness Μ AX. Here, the fine recessed portion 1 3 a formed on the top surface 3 a 1 is obtained by the following procedure. It is carried out from the center of the dielectric substrate 3 or the electrostatic chuck 1 to the outer circumference of the third or the like (four measurement images). -40-201212159 The maximum 値 among the MAX 値 of the unevenness step at each measurement position is the depth dimension of the fine concave portion 13 a formed on the top surface 3a1. Fig. 25 is a graph showing the relationship between the depth dimension of the fine concave portion 13a formed on the top surface 3a1 and the number of particles adhering to the back surface of the object to be adsorbed. The depth dimension of the concave portion 13 3 of the samples 1 to 3 was measured in accordance with the above-described photographing conditions and measurement conditions. The depth dimension of the concave portion 13a of the sample 1 is about 15 nm, the depth of the concave portion 13a of the sample 2 is about 30 nm, and the depth of the concave portion 13a of the sample 3 is about 20 nm. When the depth dimension of the recessed portion 13a is about 20 η, the number of particles adhering to the back surface of the object to be adsorbed is 600. In contrast, when the depth dimension of the concave portion 13a is 30 nm or more and 15 Onm or less, the number of particles adhering to the back surface of the object to be adsorbed can be 250 or less. Further, when the depth dimension of the concave portion 13a exceeds 15 nm, the removal of fine particles entering the inside of the concave portion 13a is difficult. Therefore, the depth of the fine recessed portion 13a is preferably 30 nm or more and 150 nm or less. (Volume Density and Alumina Content of Polycrystalline Alumina Sintered Body) In order to form the fine recessed portion 13a or the recessed portion 13b by the above-described CMP method, the bulk density and purity (content ratio) of the polycrystalline ceramic sintered body which is the underlayer are important. . Here, an example of a polycrystalline alumina sintered body will be described. -41 - 201212159 Figure 2 6 is a scanning electron micrograph of the surface of a polycrystalline alumina sintered body. Further, Fig. 26(a) shows a case where the average particle diameter of the crystal grains is 20 μηη to 50 μηι, the bulk density is 3.7, and the alumina content is 90 vvt%. Fig. 26 (b) shows a case where the average particle diameter of the crystal grains is 1·5 μm or less, the bulk density is 3.96, and the alumina content is 99.9 wt%. Comparing Fig. 26 (a) with Fig. 26 (b), it can be seen that if the bulk density is 3 · 9 6 or more and the alumina content is set to 9 9 · 9 wt % or more, the bottom layer can be increased. Since the crystalline alumina sintered body forms a dense structure, it is possible to more reliably suppress the degranulation of the crystal grains from the dielectric substrate 3. In this case, the average particle diameter of the crystal grains is set to be 0.8 μm or more, and 15 μm or less can be a dense structure. Further, a dense structure can be obtained by setting at least one of the bulk density and the purity (content ratio) within a predetermined range. However, as described above, it is preferable to set both the bulk density and the purity (content ratio) within a predetermined range. More preferably, the average particle diameter of the crystal grains is set to 〇8 μηι or more and 1.5 μηη or less. In this case, as described above, the standard deviation of the particle diameter distribution is preferably 1 μπι or less. When the polycrystalline ceramic sintered body which is the underlayer is made into a dense structure, the fine recessed portion 13a or the recessed portion 13b can be uniformly and stably formed by the above-described CMP method. As a result, the occurrence of fine particles can be greatly reduced. In this case, the bulk density can be controlled by performing ΗIP treatment (hot isostatic pressing) or the like. Further, the average particle diameter of the crystal grains can be controlled in accordance with the sintering conditions (sintering temperature, temperature history, etc.) as described above. Fig. 27 is a schematic view for explaining the number of particles - 42 - 201212159 attached to the back surface of the semiconductor wafer. In addition, Fig. 27 (a) shows the case where the polycrystalline alumina sintered body which is the bottom layer is as shown in Fig. 26 (a). Fig. 27 (b) shows the polycrystalline alumina sintered body which is the underlayer as Fig. 26 (b) In the case of Fig. 27 (a), the number of particles attached to the back surface of the 8-inch semiconductor wafer is 1058, and the case of Fig. 27(b) is attached to the back surface of the 8-inch semiconductor wafer. The number of particles is 67. (Other Embodiments of Electrostatic Suction Cup) Fig. 28 is a schematic cross-sectional view showing an electrostatic chuck 1a according to another embodiment. Fig. 28(a) is a schematic cross-sectional view for explaining an electrostatic chuck, and Fig. 28(b) is a schematic enlarged view of a portion F of Fig. 28(a). In the electrostatic chuck 1a of the present embodiment, the electrode 4 is buried inside the dielectric substrate 30. Such an electrostatic chuck 1a is produced, for example, by a green sheet printing lamination method or the like. For example, first, a tungsten paste is screen-printed from a green sheet composed of a polycrystalline ceramic sintered body (e.g., a polycrystalline alumina sintered body), whereby an electrode is formed. Then, a plurality of green sheets are pressure-bonded so that the electrodes can be embedded, and a laminate before sintering is formed. The laminated body is cut into a desired shape and sintered in a reducing atmosphere, whereby a dielectric substrate 30 in which an electrode is buried inside can be manufactured. (Manufacturing method of electrostatic chuck) -43- 201212159 Next, a method of manufacturing the electrostatic chuck according to the present embodiment will be described by way of example. Further, the dielectric substrate 3 provided on the electrostatic chuck can be manufactured as described above. Further, since a known technique can be applied to the formation, joining, mounting, and the like of each element such as the electrode 4, the description thereof will be omitted, and only the characteristic engineering will be described. Fig. 29 is a flow chart for explaining the method of manufacturing the electrostatic chuck of the embodiment. First, the main surface of the dielectric substrate 3 on which the object to be adsorbed is placed is formed into a shape of most of the projection portion 3a and the flat portion 3b by a known sandblasting method or the like. Next, as shown in Fig. 29, the projection 3a, the flat portion 3b2, the recess 13a, and the recess 13b are formed by the CMP method described above. At this time, the ratio of the defective portion is determined by the quantitative evaluation method for the defective portion, and the processing of the CMP method is continued until the ratio of the defective portion becomes a predetermined value or less. In other words, the processing of the main surface is continued until the interference pattern of the main surface obtained by the laser microscope occupies less than 1% of the area ratio. In addition, the details of the CMP method, the quantitative evaluation method for the defective portion, and the like can be the same as described above, and thus detailed description thereof will be omitted. [Industrial Applicability] As described above, according to the present invention, it is possible to provide an electrostatic chuck and a method for producing an electrostatic chuck capable of suppressing the generation of fine particles.

-44， 201212159 上的優點莫大。 【圖式簡單說明】 圖1 ( a )是用以舉例說明靜電吸盤的模式剖面圖，（ b )是（a )的A部的模式擴大圖。 圖2是用以舉例說明突起部及平面部的表面性狀、剖 面形狀等的圖表。 圖3是用以舉例說明形成於突起部的頂面之微細的凹 部的雷射顯微鏡照片。 圖4是用以舉例說明形成於平坦部之微細的凹部的掃 描型電子顯微鏡照片。 圖5是用以舉例說明將頂面設爲平坦面時的雷射顯微 鏡照片。 圖6是用以舉例說明形成於頂面3al的凹部13a的形狀 的圖，（a)是凹部13a的3次元畫像，（b) 、 （c)是用 以舉例說明凹部13a的斷面的圖。 圖7是用以舉例說明形成於平坦部3b2的凹部13b的形 狀的圖，（a)是凹部13b的3次元畫像，（b) 、（c)是 用以舉例說明凹部13b的斷面的圖。 圖8是用以舉例說明開口於平面部的孔的深度尺寸的 圖表。 圖9是用以舉例說明微細的凹部的長度測定的雷射顯 微鏡照片。 圖1 0是用以舉例說明在多結晶陶瓷燒結體的表面顯現 •45- 201212159 之結晶粒的長度測定的雷射顯微鏡照片。 圖11是用以舉例說明在介電質基板的表面區域產生的 龜裂的掃描型電子顯微鏡照片。 圖12是用以舉例說明表面區域的一部分似脫離的狀態 的掃描型電子顯微鏡照片。 圖1 3 ( a )是用以舉例說明在內部存在缺陷部的部分 產生的干涉條紋的雷射顯微鏡照片，（b )是（a )的B-B 線剖面的掃描型電子顯微鏡（SEM; Scanning Electron Microscope)照片，（c)是（b)的D部的擴大照片，（d )是與（a)同部分的掃描型電子顯微鏡照片。 圖14 (a)是用以舉例說明在內部存在缺陷部的部分 產生的干涉條紋的雷射顯微鏡照片，（b )是（a)的C-C 線剖面的掃描型電子顯微鏡照片。 圖1 5是用以舉例說明被2値化處理的畫像的照片。 圖16是用以舉例說明利用CMP法來除去缺陷部的狀態 的圖表。 圖17是用以舉例說明利用CMP法來除去缺陷部之前的 狀態的圖表。 圖18是用以舉例說明利用CMP法來除去缺陷部的狀態 的圖表。 圖19是用以舉例說明利用CMP法來除去缺陷部之前的 狀態的圖表。 圖20 ( a)是結晶粒的平均粒子徑爲1.8 μηι程度的情況 ，（b )是結晶粒的平均粒子徑爲丨.4μηι程度的情況。 -46- 201212159 圖2 1是用以舉例說明藉由雷射顯微鏡所攝取的多結晶 陶瓷燒結體的照片。 圖22是用以舉例說明結晶粒的平均粒子徑與粒子徑分 布的標準偏差的圖表。 圖23是用以舉例說明結晶粒的平均粒子徑與粒子徑分 布的標準偏差的圖表。 圖24是用以舉例說明微細的凹部的深度測定的圖，（ a)是用以舉例說明測定値的分布的圖表，（b)是用以舉 例說明測定位置的雷射顯微鏡照片。 圖25是表示形成於頂面3al的微細的凹部13a的深度尺 寸與附著於被吸附物的背面的微粒數的關係圖表》 圖26(3)是結晶粒的平均粒子徑爲2(^111〜5(^111，體 積密度爲3.7，氧化鋁含有率爲90wt%的情況，（b )是結 晶粒的平均粒子徑爲1.5μπι以下，體積密度爲3.96，氧化 鋁含有率爲99.9wt%的情況。 圖27是用以舉例說明附著於半導體晶圓的背面的微粒 數的模式圖。另外，（a)是成爲底層的多結晶氧化鋁燒 結體爲圖26(a)所示者的情況，（b)是成爲底層的多結 晶氧化鋁燒結體爲圖26 ( b )所示者的情況。 圖2 8 ( a )是用以舉例說明其他的實施形態的靜電吸 盤的模式剖面圖，（b )是（a )的F部的模式擴大圖。 圖2 9是用以舉例說明靜電吸盤的製造方法的流程圖。 【主要元件符號說明】 -47- 201212159 1 :靜電吸盤 1 a :靜電吸盤 2 :基台 3 :介電質基板 3 a :突起部 3 a 1 :頂面 3 b :平面部 3bl :孔 3 b 2 :平坦部 3 c :空間 4 :電極 1 0a :電源 1 0 b :電源 1 1 :貫通孔 13a :凹部 1 3 b :凹部 30 :介電質基板 -48The advantages on -44, 201212159 are great. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1(a) is a schematic cross-sectional view for illustrating an electrostatic chuck, and Fig. 1(b) is a schematic enlarged view of a portion A of (a). Fig. 2 is a graph for exemplifying the surface properties, the cross-sectional shape, and the like of the projections and the flat portions. Fig. 3 is a laser micrograph for illustrating a fine concave portion formed on the top surface of the projection. Fig. 4 is a scanning electron micrograph showing an example of a fine concave portion formed in a flat portion. Fig. 5 is a photograph of a laser microscope for illustrating a case where the top surface is a flat surface. Fig. 6 is a view for exemplifying the shape of the concave portion 13a formed on the top surface 3a, (a) is a three-dimensional image of the concave portion 13a, and (b) and (c) are diagrams for illustrating a cross section of the concave portion 13a. . Fig. 7 is a view for exemplifying the shape of the concave portion 13b formed in the flat portion 3b2, wherein (a) is a three-dimensional image of the concave portion 13b, and (b) and (c) are views for illustrating a cross section of the concave portion 13b. . Fig. 8 is a graph for illustrating the depth dimension of a hole opened in a plane portion. Fig. 9 is a photograph of a laser microscope for illustrating the measurement of the length of a fine recess. Fig. 10 is a laser microscope photograph for illustrating the measurement of the length of the crystal grains of the surface of the polycrystalline ceramic sintered body from ?45 to 201212159. Fig. 11 is a scanning electron micrograph showing an example of cracks generated in a surface region of a dielectric substrate. Fig. 12 is a scanning electron micrograph showing a state in which a part of the surface region appears to be detached. Fig. 1 3 (a) is a laser micrograph for illustrating the interference fringes generated in the portion where the defect portion is present, and (b) is a scanning electron microscope (SEM; Scanning Electron Microscope) of the BB line cross section of (a) The photograph, (c) is an enlarged photograph of the D portion of (b), and (d) is a scanning electron microscope photograph of the same portion as (a). Fig. 14 (a) is a laser micrograph showing an example of interference fringes generated in a portion where a defect portion is present, and (b) is a scanning electron micrograph of a cross section taken along line C-C of (a). Fig. 15 is a photograph for illustrating an image processed by the sputum. Fig. 16 is a graph for explaining a state in which a defective portion is removed by a CMP method. Fig. 17 is a graph for explaining a state before the defective portion is removed by the CMP method. Fig. 18 is a graph for explaining a state in which a defective portion is removed by a CMP method. Fig. 19 is a graph for explaining a state before the defective portion is removed by the CMP method. Fig. 20 (a) shows a case where the average particle diameter of the crystal grains is about 1.8 μη, and (b) is a case where the average particle diameter of the crystal grains is about 0.4 μm. -46- 201212159 Fig. 2 1 is a photograph for illustrating a polycrystalline ceramic sintered body taken by a laser microscope. Fig. 22 is a graph for exemplifying the standard deviation of the average particle diameter of the crystal grains and the particle diameter distribution. Fig. 23 is a graph for illustrating the standard deviation of the average particle diameter of the crystal grains and the particle diameter distribution. Fig. 24 is a view for exemplifying the depth measurement of the fine concave portion, (a) is a graph for exemplifying the measurement of the distribution of ruthenium, and (b) is a laser microscope photograph for explaining the measurement position. Fig. 25 is a graph showing the relationship between the depth dimension of the fine concave portion 13a formed on the top surface 3a1 and the number of particles adhering to the back surface of the object to be adsorbed. Fig. 26 (3) shows that the average particle diameter of the crystal grains is 2 (^111~ 5 (^111, the bulk density is 3.7, the alumina content is 90% by weight, (b) is the case where the average particle diameter of the crystal grains is 1.5 μm or less, the bulk density is 3.96, and the alumina content is 99.9 wt%. Fig. 27 is a schematic view for exemplifying the number of particles adhering to the back surface of the semiconductor wafer. (a) The case where the polycrystalline alumina sintered body to be the underlayer is as shown in Fig. 26 (a), b) is a case where the polycrystalline alumina sintered body to be the bottom layer is as shown in Fig. 26 (b). Fig. 2 (a) is a schematic sectional view for explaining an electrostatic chuck of another embodiment, (b) It is a mode enlargement diagram of the F part of (a). Fig. 29 is a flowchart for explaining the manufacturing method of the electrostatic chuck. [Description of main component symbols] -47- 201212159 1 : Electrostatic chuck 1 a : Electrostatic chuck 2 : Abutment 3: dielectric substrate 3 a : protrusion 3 a 1 : top surface 3 b : flat portion 3b l : hole 3 b 2 : flat portion 3 c : space 4 : electrode 1 0a : power supply 1 0 b : power supply 1 1 : through hole 13a : recess 1 3 b : recess 30 : dielectric substrate -48

Claims (1)

201212159 七、申請專利範圍： 1. 一種靜電吸盤，係具備介電質基板，該介電質基板 係具有：形成於載置被吸附物側的主面之突起部、及形成 於上述突起部的周邊之平面部，其特徵爲： 上述介電質基板係由多結晶陶瓷燒結體所形成， 利用雷射顯微鏡所求取之上述主面的干涉條紋佔有面 積率爲未滿1 %。 2. 如申請專利範圍第1項之靜電吸盤，其中，上述多 結晶陶瓷燒結體的結晶粒的平均粒子徑係比上述突起部的 高度尺寸更小。 3. 如申請專利範圍第2項之靜電吸盤，其中，上述平 均粒子徑爲1.5 μιη以下。 4. 如申請專利範圍第2項之靜電吸盤，其中，上述結 晶粒的粒子徑分布的標準偏差爲1 μηι以下。 5·如申請專利範圍第1項之靜電吸盤，其中，上述介 電質基板係由多結晶氧化鋁燒結體所形成，體積密度爲 3.96以上。 6. 如申請專利範圍第1項之靜電吸盤，其中，上述介 電質基板係由多結晶氧化鋁燒結體所形成，氧化錨含有·率 爲99.9wt%以上。 7. 如申請專利範圍第1項之靜電吸盤，其中，上述介 電質基板係體積電阻率在靜電吸盤的使用溫度區域φ爲 1 08 Ω c m 以上、1 〇13 Ω c m 以下。 8. 如申請專利範圍第7項之靜電吸盤，其中，上述介 -49- 201212159 電質基板係由多結晶氧化鋁燒結體所形成，氧化鋁含有率 爲99.4wt%以上。 9. 一種靜電吸盤的製造方法，該靜電吸盤係具備介電 質基板，該介電質基板係具有：形成於載置被吸附物側的 主面之突起部、及形成於上述突起部的周邊之平面部，其 特徵爲= 上述介電質基板係由多結晶陶瓷燒結體所形成， 上述主面的加工係繼續至利用雷射顯微鏡所求取之上 述主面的干涉條紋佔有面積率爲未滿1 %。 -50-201212159 VII. Patent application scope: 1. An electrostatic chuck comprising a dielectric substrate having a protrusion formed on a main surface on which an object to be adsorbed is placed, and a protrusion formed on the protrusion The peripheral planar portion is characterized in that the dielectric substrate is formed of a polycrystalline ceramic sintered body, and the interference fringe occupying area ratio of the main surface obtained by a laser microscope is less than 1%. 2. The electrostatic chuck according to claim 1, wherein the crystal grain of the polycrystalline ceramic sintered body has an average particle diameter smaller than a height dimension of the protrusion. 3. The electrostatic chuck according to item 2 of the patent application, wherein the average particle diameter is 1.5 μm or less. 4. The electrostatic chuck according to item 2 of the patent application, wherein the standard deviation of the particle diameter distribution of the above-mentioned crystal grains is 1 μηι or less. 5. The electrostatic chuck according to claim 1, wherein the dielectric substrate is formed of a polycrystalline alumina sintered body having a bulk density of 3.96 or more. 6. The electrostatic chuck according to claim 1, wherein the dielectric substrate is formed of a polycrystalline alumina sintered body, and an oxidation anchor content of 99.9 wt% or more. 7. The electrostatic chuck according to claim 1, wherein the dielectric substrate has a volume resistivity of 1 08 Ω c m or more and 1 〇 13 Ω c m or less in the use temperature region φ of the electrostatic chuck. 8. The electrostatic chuck according to claim 7, wherein the dielectric substrate is formed of a polycrystalline alumina sintered body, and the alumina content is 99.4% by weight or more. 9. A method of manufacturing an electrostatic chuck, comprising: a dielectric substrate having: a protrusion formed on a main surface on which an object to be adsorbed is placed; and a periphery formed on the protrusion The planar portion is characterized in that the dielectric substrate is formed of a polycrystalline ceramic sintered body, and the processing of the main surface continues until the interference fringe area ratio of the main surface obtained by the laser microscope is not Full 1%. -50-
TW100128677A 2010-08-11 2011-08-11 Electrostatic chuck and production method thereof TW201212159A (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2010180605 2010-08-11
JP2011171019A JP5510411B2 (en) 2010-08-11 2011-08-04 Electrostatic chuck and method for manufacturing electrostatic chuck

Publications (1)

Publication Number Publication Date
TW201212159A true TW201212159A (en) 2012-03-16

Family

ID=45567792

Family Applications (1)

Application Number Title Priority Date Filing Date
TW100128677A TW201212159A (en) 2010-08-11 2011-08-11 Electrostatic chuck and production method thereof

Country Status (6)

Country Link
US (1) US8971010B2 (en)
JP (1) JP5510411B2 (en)
KR (1) KR101450025B1 (en)
CN (1) CN103053018A (en)
TW (1) TW201212159A (en)
WO (1) WO2012020832A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103681433A (en) * 2012-08-29 2014-03-26 Toto株式会社 Electrostatic chuck

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
SG171819A1 (en) * 2008-12-25 2011-07-28 Ulvac Inc Method of manufacturing chuck plate for use in electrostatic chuck
KR101173578B1 (en) * 2012-03-09 2012-08-13 윈텍 주식회사 Electrostatic attraction type inspection table for electric parts
JP5441020B1 (en) * 2012-08-29 2014-03-12 Ｔｏｔｏ株式会社 Electrostatic chuck
US20150062772A1 (en) * 2013-08-27 2015-03-05 Varian Semiconductor Equipment Associates, Inc Barrier Layer For Electrostatic Chucks
JP6283532B2 (en) * 2014-02-26 2018-02-21 東京エレクトロン株式会社 Manufacturing method of electrostatic chuck
WO2019065726A1 (en) * 2017-09-28 2019-04-04 京セラ株式会社 Substrate for mounting light-emitting element, circuit substrate for mounting light-emitting element comprising same, and light-emitting element module
WO2020261990A1 (en) * 2019-06-28 2020-12-30 日本碍子株式会社 Wafer mounting stand and method for manufacturing same
CN113053774A (en) * 2019-12-27 2021-06-29 迪科特测试科技(苏州)有限公司 Probe apparatus

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5531835A (en) * 1994-05-18 1996-07-02 Applied Materials, Inc. Patterned susceptor to reduce electrostatic force in a CVD chamber
US5548470A (en) * 1994-07-19 1996-08-20 International Business Machines Corporation Characterization, modeling, and design of an electrostatic chuck with improved wafer temperature uniformity
JP4094262B2 (en) * 2001-09-13 2008-06-04 住友大阪セメント株式会社 Adsorption fixing device and manufacturing method thereof
KR100719185B1 (en) 2002-01-31 2007-05-16 엔지케이 스파크 플러그 캄파니 리미티드 Ceramic sintered body and process for producing the same
US7072166B2 (en) * 2003-09-12 2006-07-04 Axcelis Technologies, Inc. Clamping and de-clamping semiconductor wafers on a J-R electrostatic chuck having a micromachined surface by using force delay in applying a single-phase square wave AC clamping voltage
KR100666039B1 (en) * 2003-12-05 2007-01-10 동경 엘렉트론 주식회사 Electrostatic chuck
US7672110B2 (en) * 2005-08-29 2010-03-02 Applied Materials, Inc. Electrostatic chuck having textured contact surface
JP2007214288A (en) * 2006-02-08 2007-08-23 Toto Ltd Electrostatic chuck
JP2007255957A (en) * 2006-03-22 2007-10-04 Nikon Corp Inspection method of wafer chuck
GB0611156D0 (en) * 2006-06-07 2006-07-19 Qinetiq Ltd Optical inspection
JP4890421B2 (en) 2006-10-31 2012-03-07 太平洋セメント株式会社 Electrostatic chuck
JP2008177339A (en) 2007-01-18 2008-07-31 Sumitomo Osaka Cement Co Ltd Electrostatic chuck
JP2009060035A (en) * 2007-09-03 2009-03-19 Shinko Electric Ind Co Ltd Electrostatic chuck member, its manufacturing method, and electrostatic chuck apparatus

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103681433A (en) * 2012-08-29 2014-03-26 Toto株式会社 Electrostatic chuck
CN103681433B (en) * 2012-08-29 2018-01-30 Toto株式会社 Electrostatic chuck

Also Published As

Publication number Publication date
US8971010B2 (en) 2015-03-03
WO2012020832A1 (en) 2012-02-16
CN103053018A (en) 2013-04-17
US20130201598A1 (en) 2013-08-08
KR101450025B1 (en) 2014-10-13
JP5510411B2 (en) 2014-06-04
KR20130030292A (en) 2013-03-26
JP2012060106A (en) 2012-03-22

Similar Documents

Publication Publication Date Title
TW201212159A (en) Electrostatic chuck and production method thereof
TWI434370B (en) Electrostatic sucker
TWI386984B (en) Silicon electrode assembly surface decontamination by acidic solution
KR102098926B1 (en) Rare-earth oxide based erosion resistant coatings for semiconductor application
JP4744855B2 (en) Electrostatic chuck
TWI427735B (en) Electrostatic chuck and electrostatic chuck manufacturing method
JP5537001B2 (en) Surface-treated ceramic member, manufacturing method thereof, and vacuum processing apparatus
JP2008091353A (en) Electrostatic chuck
TWI724797B (en) Member for semiconductor manufacturing device, semiconductor manufacturing device and display manufacturing device including the member for semiconductor manufacturing device
JP2007201363A (en) Member for vacuum suction of semiconductor wafer
JP2010109106A (en) Suction cup and vacuum suction apparatus
JP4006535B2 (en) Semiconductor or liquid crystal manufacturing apparatus member and manufacturing method thereof
KR101974386B1 (en) Electrode static chuck
KR100989230B1 (en) Electrostatic chuck
JP2012060107A (en) Surface evaluation method of attraction retainer
JP7308955B2 (en) Clamp jigs and cleaning equipment
JP7284553B2 (en) Substrate with thermal spray coating and method for manufacturing the same
WO2023127796A1 (en) Clamping jig and cleaning device
JP2007217774A (en) Plasma-resistant thermal-sprayed member
JP2011162422A (en) Surface-treated ceramic member and method for manufacturing the same